Treatment of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults

INTRODUCTION — Axial spondyloarthritis (SpA), which includes ankylosing spondylitis (AS) and nonradiographic axial SpA (nr-axSpA), is a chronic inflammatory condition manifested by back pain and progressive spinal stiffness. AS and nr-axSpA differ in that significant abnormalities of affected sacroiliac joints are observed by conventional radiography in patients with AS but not (yet) in those with nr-axSpA.

Axial SpA characteristically presents in young adults with a peak age of onset between 20 and 30 years. Although primarily thought of as a spinal disease, enthesitis, as well as arthritis of peripheral joints, which is sometimes transient, occur in up to 50 percent of patients. In addition, other organs such as the eyes, bowel, lungs, heart, and kidneys can be affected.

The treatment, monitoring, and prognosis of axial SpA in adults are presented here. The clinical manifestations and diagnosis of axial and peripheral SpA in adults, the treatment of peripheral SpA, and a detailed discussion of SpA in children are presented separately. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults" and "Diagnosis and differential diagnosis of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults" and "Clinical manifestations and diagnosis of peripheral spondyloarthritis in adults" and "Treatment of peripheral spondyloarthritis" and "Spondyloarthritis in children".)

GOALS AND GENERAL PRINCIPLES OF MANAGEMENT — The primary goals of management for patients with axial spondyloarthritis (SpA) are to optimize short- and long-term health-related quality of life through the following [1]:

●Relief of symptoms – To eliminate symptoms such as pain, stiffness, and fatigue or to reduce them to the minimal possible level

●Maintenance of function – To maintain the best possible functional capacity

●Prevention of complications of spinal disease – To prevent limitation in ranges of motion and flexion contractures, especially dorsal kyphosis

●Minimization of extraspinal and extraarticular manifestations and comorbidities – To reduce the impact of axial SpA-associated disorders such as uveitis and aortic valve insufficiency

●Maintenance of effective psychosocial functioning – To preserve social participation, prevent job loss, and improve health status and function

General principles of management and approach to therapy in axial SpA include [1-4]:

●Most patients benefit from care by an expert in rheumatologic disease, such as a rheumatologist, and care should be coordinated with appropriate clinical specialists, depending upon the clinical features, such as a dermatologist for psoriasis, a gastroenterologist for inflammatory bowel disease (IBD), and an ophthalmologist for uveitis.

●Disease activity should be regularly measured and therapy adjusted accordingly to improve outcome.

●All patients should receive nonpharmacologic measures, including patient education, physical therapy and exercise, encouragement to participate in support groups, and encouragement and support of smoking cessation. (See 'Nonpharmacologic interventions' below.)

●Pharmacologic treatment of the axial and peripheral articular manifestations is very helpful in reducing symptoms and preventing functional limitations that can result directly from disease activity. Pharmacotherapy includes one or more of the following: nonsteroidal antiinflammatory drugs (NSAIDs), non-NSAID analgesics, nonbiologic (conventional synthetic [cs]) disease-modifying antirheumatic drugs (DMARDs), and biologic (b)DMARDs. Unlike rheumatoid arthritis, oral (low-dose) glucocorticoids have no role in axial SpA, but intraarticular injections may be helpful to some patients.

The choice of therapy is based upon the selection of agents that will be effective alone or in combination for the clinical manifestations that are present in a given patient. Most of the treatments for the different clinical manifestations overlap, but some are more effective for one or another feature. Treatment choices, particularly the use of bDMARDs such as a tumor necrosis factor (TNF) inhibitors and interleukin (IL) 17 inhibitors, may also be influenced by the presence of findings of another disease associated with axial SpA that may require one of these agents, such as psoriasis, IBD, and uveitis.

●In addition to control of disease activity, symptoms that require recognition and appropriate treatment include anxiety, depression, fatigue, sleep disturbance, and helplessness, which also contribute to functional limitations in some patients with axial SpA [5].

Active patient engagement in shared decision-making with their clinical team regarding their treatment is important in the management of axial SpA, which is usually lifelong. Patients vary greatly in their disease pattern, response to different therapies, rate of disease progression, and goals and preferences.

INITIAL THERAPY — The initial treatment interventions for most patients with axial spondyloarthritis (SpA; ankylosing spondylitis [AS] and nonradiographic axial SpA [nr-axSpA]) include a series of nonpharmacologic measures (see 'Nonpharmacologic interventions' below) and nonsteroidal antiinflammatory drug (NSAID) therapy. (See 'Initial drug therapy with NSAIDs' below.)

Nonpharmacologic interventions — All patients newly diagnosed with axial SpA should receive the following initial and ongoing interventions:

●Patient education – Patients should receive education about the nature of their disease and advice about the need for a lifelong exercise and posture-training program and about their working and leisure habits relevant to axial SpA. Patients should also be educated about the importance of regular follow-up and management of comorbidities. Any patients receiving pharmacologic treatment should be instructed about their medications, the need for adhering to regular drug administration, and the monitoring of disease activity and for potential side effects of therapies.

●Counseling regarding smoking cessation – Smoking cessation should be advised because cigarette smoking has an adverse effect on SpA, especially with respect to greater progression of structural damage of the spine, in addition to its adverse effects upon cardiovascular risk and other aspects of health [6]. (See "Overview of smoking cessation management in adults".)

●Depression screening and psychosocial support – Patients should undergo screening for anxiety and depression and be encouraged to participate in patient support groups and arthritis self-help programs [7]. (See "Screening for depression in adults".)

●Exercises and physical therapy – Exercise improves cardiovascular fitness and the disease activity of SpA patients [8,9]. Patients newly diagnosed with axial SpA should be referred to a physical therapist for an initial evaluation and training. Exercises include postural training, range of motion exercises, stretching, recreational activities, and sometimes hydrotherapy. Spinal manipulation should be avoided in patients with spinal fusion or advanced spinal osteoporosis [4].

Home exercises are effective, but supervised exercise programs or formal physical therapy can be of greater benefit, and exercise therapy combined with hydrotherapy might be more effective than exercises alone [10]. Inpatient rehabilitation is rarely needed but may be a solution for those patients with concomitant psychosocial problems or for encouraging work reintegration [11].

Even those patients who are doing well clinically with pharmacologic treatment will also benefit from education and exercise [12-15]. A randomized trial involving 62 patients clinically stable on anti-tumor necrosis factor (TNF) therapy showed statistically significant benefit in the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and in spinal range of motion after two and six months from the combination of an intensive rehabilitation exercise treatment program with an educational-behavioral program, compared with the educational-behavioral treatment alone or the results of a control group [12].

Additional instructional materials for exercising are available from patient support organizations, such as the Spondylitis Association of America and the National Ankylosing Spondylitis Society in the United Kingdom, including guidebooks and audio and video aids, such as video demonstrations of exercises for axial SpA [16].

Initial drug therapy with NSAIDs — In most patients with symptomatic axial SpA, we suggest a nonsteroidal antiinflammatory drug (NSAID) as initial therapy. Examples include naproxen (up to 500 mg twice daily), celecoxib (up to 200 mg twice daily), or ibuprofen (up to 800 mg three times daily), although any non-aspirin NSAID may be effective (table 1). Regardless of the NSAID used, the maximum dose is often required. The potential gastrointestinal, renal, cardiovascular, and other risks of NSAIDs need to be considered when using these agents. The potential adverse effects of NSAIDs are reviewed in detail separately. (See "Nonselective NSAIDs: Overview of adverse effects" and "Overview of COX-2 selective NSAIDs", section on 'Toxicities and possible toxicities'.)

Each NSAID tried should be assessed for its efficacy in alleviating symptoms such as pain and stiffness at a full antiinflammatory dose on a regular continuing basis for at least two to four weeks before switching to a second NSAID [17].

In some patients, NSAIDs are the only medications required. Approximately 70 to 80 percent of AS patients report substantial relief of their symptoms, including back pain and stiffness, with NSAIDs [18]. By comparison, this is significantly more than the 15 percent observed in patients with chronic mechanical low back pain. NSAIDs are also helpful for reducing the symptoms of peripheral arthritis [19]. In an observational study involving patients with high disease activity prior to starting NSAIDs, after four weeks of NSAID therapy, an Assessment of SpondyloArthritis International Society (ASAS) 40 percent (ASAS40) response was experienced by 35 percent, and a partial remission was achieved by 16 percent [20].

There is no evidence that one NSAID is more effective than another in axial SpA [4,21-23]. A 2015 systematic review and meta-analysis of multiple randomized trials showed benefits from both nonselective and cyclooxygenase-2 (COX-2) selective NSAIDs and little evidence that harm from NSAID use differed from placebo after 12 weeks of therapy [22]. The use of NSAIDs in axial SpA is also consistent with expert guidelines of major organizations [1,2].

One case-control study found that current use of diclofenac, compared with remote use, was associated with an elevated risk of myocardial infarction in patients with SpA (odds ratio [OR] 3.32, 95% CI 1.57-7.03) [24]. There was also evidence of increased cardiovascular risk in patients with osteoarthritis (OA), but the increased risk with diclofenac was greater in patients with SpA compared with that seen with OA (OR 2.64, 95% CI 1.24-5.58). Such risk was not seen in patients on naproxen. Cardiovascular risk associated with diclofenac and with other NSAIDs is reviewed in detail separately. (See "NSAIDs: Adverse cardiovascular effects".)

Duration of NSAID therapy — Once a particular nonsteroidal antiinflammatory drug (NSAID) has been determined to be effective within two to four weeks, it is usually then used on demand (as needed) according to symptoms, although some patients require ongoing daily therapy to maintain benefit. Some data have suggested that treating patients continuously with NSAIDs even though patients are asymptomatic might prevent formation of growth of syndesmophytes in the spine [25]. However, this could not be verified in a subsequent study [26]. We do not use NSAIDs continuously in asymptomatic patients [27].

Prevention of syndesmophyte formation was suggested in a 2005 trial involving 215 patients treated with celecoxib (100 to 200 mg twice daily) either continuously or as needed, those receiving continuous therapy exhibited less radiographic progression in the spine at two years [25]. By contrast, another randomized trial, in which 167 patients with AS were treated with diclofenac either continuously or as needed, found no evidence at two years for reduced radiographic change in the group assigned to continuous therapy [26].

Once treatment has been initiated, the disease activity should be monitored regularly, and the degree of response should be assessed. (See 'Evaluation and monitoring' below.)

INADEQUATE RESPONSE TO NSAIDs — In patients with symptoms due to active axial spondyloarthritis (SpA) and an inadequate response to initial therapy with two different nonsteroidal antiinflammatory drugs (NSAIDs), we suggest a tumor necrosis factor (TNF) inhibitor rather than continuing NSAIDs alone or conventional synthetic (cs) disease-modifying antirheumatic drugs (DMARDs) such as sulfasalazine (SSZ). The NSAIDs should have been used consecutively in an adequate dose for at least two to four weeks each. Any of the TNF inhibitors is an acceptable option (see 'TNF inhibitors' below). The choice between them is based upon patient preferences regarding the route and frequency of administration, clinician preference and experience, regulatory and cost constraints, and possible coexisting conditions such as psoriasis, inflammatory bowel disease (IBD), or uveitis.

Several classes of DMARDs may be effective in patients with an inadequate response to NSAIDs. These include:

●TNF inhibitors, which include etanercept, infliximab, adalimumab, golimumab, certolizumab pegol, and biosimilars of these agents. (See 'TNF inhibitors' below.)

●Anti-interleukin (IL) 17 antibodies, which include secukinumab and ixekizumab. (See 'Interleukin 17 inhibitors' below.)

●Janus kinase (JAK) inhibitors, which include tofacitinib and upadacitinib. (See 'Janus kinase inhibitors' below.)

Monoclonal antibodies against TNF (eg, infliximab, adalimumab, certolizumab pegol, golimumab, and their biosimilars) are preferred for patients with IBD or uveitis. Anti-IL-17 antibodies are preferred for patients with significant psoriasis [27].

Unlike in, for instance, rheumatoid arthritis, none of the agents in the three categories noted need to be used together with a cs immunosuppressive agent (DMARD) such as methotrexate (MTX), as there is no evidence this offers greater benefit. (See 'Efficacy of TNF inhibitors' below.)

The use of these agents for active axial SpA is particularly appropriate for those with high or very high disease activity, although they are effective in patients with mild to moderate symptoms as well (see 'Efficacy of TNF inhibitors' below). The high cost of these therapies and efforts by regulators or payers to control such costs are major impediments to wider availability of these agents for use in symptomatic SpA [28]. High disease activity can be defined as a Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) ≥4.0 or Ankylosing Spondylitis Disease Activity Score (ASDAS) ≥2.1 [17]. (See 'Evaluation and monitoring' below.)

Before proceeding with one of these agents, it is also important to clinically exclude other causes of the patient's symptoms, including fibromyalgia, which is present in a substantial minority of patients with axial SpA and can mimic symptoms of SpA as well [29-32]. The differential diagnosis of SpA and the diagnosis and treatment of fibromyalgia are described separately. (See "Diagnosis and differential diagnosis of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Differential diagnosis' and "Clinical manifestations and diagnosis of fibromyalgia in adults" and "Treatment of fibromyalgia in adults".)

In patients with symptoms and findings predominantly of peripheral arthritis, we use a csDMARD first before switching to a TNF inhibitor. Use of SSZ prior to TNF inhibitors is limited to patients with prominent peripheral arthritis [4]. (See 'Peripheral arthritis and periarticular disease' below.)

TNF inhibitors

Use of TNF inhibitors

●Dosing – The typical doses of the TNF inhibitors for both ankylosing spondylitis (AS) and nonradiographic axial SpA (nr-axSpA) are:

•Etanercept – Usually administered at a dose of 50 mg once weekly as a subcutaneous injection; it is occasionally given as an injection of 25 mg twice a week

•Infliximab – 5 mg/kg by intravenous infusion at zero, two, and six weeks followed by a maintenance dose of 5 mg/kg every six to eight weeks

•Adalimumab – 40 mg by subcutaneous injection every two weeks

•Golimumab – 50 mg by subcutaneous injection every four weeks

•Certolizumab pegol – 400 mg by subcutaneous injection at zero, two, and four weeks, followed by 200 mg every other week or 400 mg every four weeks

There is little information on whether higher doses of any of these TNF antagonists would confer greater efficacy [33].

Biosimilar agents (eg, for infliximab, etanercept, and adalimumab) have the same dosing regimen as originators and seem to have a similar efficacy at lower costs [34,35]. (See "Overview of biologic agents in the rheumatic diseases", section on 'Biosimilars for biologic agents'.)

●Contraindications – The contraindications to TNF inhibitor use are the same as those for use in other diseases, such as rheumatoid arthritis. Briefly summarized, these include:

•Active infection

•Latent (untreated) tuberculosis (TB)

•Demyelinating disease (eg, multiple sclerosis, optic neuritis)

•Heart failure

•Malignancy

All patients should be tested for latent TB prior to starting TNF inhibitors, anti-IL-17 antibodies, and JAK inhibitors.

TNF inhibitors are probably safe in patients who are pregnant or breastfeeding, but the evidence is limited. The risks of these biologics in general and also individually are discussed in detail separately. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Tumor necrosis factor inhibitors'.)

The safety of these agents and risk of recurrent malignancy in patients with a history of prior malignancy is less well established than in patients without such a history; the available data are discussed in detail separately. (See "Tumor necrosis factor-alpha inhibitors: Risk of malignancy".)

●Adverse effects – The risks and adverse effects of the TNF inhibitors are described in detail separately. In general, the risk of serious adverse effects is small [36]. (See "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects" and "Tumor necrosis factor-alpha inhibitors: Bacterial, viral, and fungal infections" and "Risk of mycobacterial infection associated with biologic agents and JAK inhibitors" and "Tumor necrosis factor-alpha inhibitors: Risk of malignancy" and "Tumor necrosis factor-alpha inhibitors: Induction of antibodies, autoantibodies, and autoimmune diseases".)

●Monitoring – (See 'Evaluation and monitoring' below.)

●Duration of therapy – (See 'Duration and tapering of therapy with TNF inhibitors' below.)

Efficacy of TNF inhibitors — The ability of the TNF inhibitors to reduce disease activity in patients with axial SpA, including both AS and nr-axSpA, has been demonstrated in multiple randomized trials and several meta-analyses of randomized trials [37-39]. Trials have been performed with each of the five widely available TNF inhibitors (ie, infliximab, etanercept, adalimumab, certolizumab, and golimumab), as well as a number of biosimilars to these drugs, which are becoming increasingly available in the United States, Europe, and other regions worldwide. (See "Overview of biologic agents in the rheumatic diseases", section on 'Biosimilars for biologic agents'.)

As examples of the available evidence:

●The efficacy of TNF inhibitors (including etanercept, adalimumab, infliximab, certolizumab, and golimumab) in the treatment of axial SpA was documented in a 2015 systematic review and meta-analysis of randomized trials involving over 2400 patients in which substantial improvements in disease activity and function were seen [37]. Patients treated with these agents, compared with placebo, were significantly more likely to achieve at least 40 percent improvement from baseline, measured with the Assessment of SpondyloArthritis International Society (ASAS) 40 percent (ASAS40) composite response measure (odds ratio [OR] 4.73, 95% CI 3.75-5.98).

The ASAS response is a composite measure, which includes:

•Patient global assessment

•Patient assessment of pain

•A functional assessment, such as ability to carry out certain activities with which the patient had difficulty prior to treatment

•Degree of inflammation as assessed by morning stiffness

●A 2007 meta-analysis indicated that all three of the TNF inhibitors then available (adalimumab, etanercept, and infliximab) were similar in efficacy in patients with AS [40]. At week 12 of trials, patients treated with the TNF inhibitors were 3.6-fold more likely, compared with those treated with placebo, to achieve substantial clinical improvement using a composite measure [41]. There is a lack of head-to-head studies comparing the efficacy of these TNF inhibitors. Indirect comparison in a 2021 review of network meta-analyses and treatment comparisons suggested that groups on infliximab and golimumab have higher probability of responses at 12 weeks [42].

The clinical responses are typically rapid. Eighty percent of patients who experienced a >50 percent response according to the BASDAI (see 'Evaluation and monitoring' below) by 12 weeks did so within the first six weeks of treatment. The number needed to treat (NNT) to achieve a partial remission is estimated to be from 2.3 to 6 [43]. In patients with advanced AS, treatment with TNF inhibitors was also found to be effective [44]. The long-term benefit of TNF inhibitor therapy appears to be durable [45].

Patients with a shorter duration of disease, and to a lesser extent those with elevated C-reactive protein (CRP) levels or younger age, are most likely to have a good response [46-48]. Improvement of greater than 50 percent may be seen in up to 80 percent of such patients [49].

TNF inhibitors are also effective in patients with clinically symptomatic nr-axSpA who failed treatment with NSAIDs [50-57].

Concomitant use of a csDMARD, such as MTX, is not required as most of the limited evidence indicates that this provides no additional benefit in patients with SpA but may increase the cost and the risk of adverse effects [58-61].

Pooled data from 12 European registries, including a total of 24,195 axial SpA patients, found that the disease becomes inactive in routine care after six months of treatment with a first TNF inhibitor in approximately one-fourth of treated AS patients and one-fifth of nr-axSpA patients, with drug retention rates at 12 months of 80 and 73 percent, respectively [62].

The frequency of an inadequate response to the initial TNF inhibitor has also been analyzed in the United States using an insurance claims database; of patients newly started on a TNF inhibitor from 2009 to 2013, only approximately one-third of males and less than one-fourth of females with AS continued to use the initial TNF inhibitor over a two-year follow-up period [63].

Duration and tapering of therapy with TNF inhibitors — For almost all patients, we continue treatment with TNF inhibitors indefinitely without discontinuing. However, tapering TNF inhibitors may be considered in patients who have remained in clinical remission for at least six months. For such patients, we have a detailed discussion that the risks of relapse vary among patients. Although some patients will tolerate TNF inhibitor tapering, we are generally reluctant to discontinue a TNF inhibitor completely because of the high risk of relapse [64]. The strategy for tapering IL-17 inhibitors is described in the section on ixekizumab. (See 'Ixekizumab' below.)

●Tapering therapy – Multiple studies suggest that some patients will be able to maintain sustained low disease activity after dose reduction of TNF inhibitor therapy [53,59,65-72]. In the Study to Evaluate Maintenance of Sustained Remission of axial SpA With Certolizumab Compared to Placebo (C-OPTIMISE) trial, patients with early axial SpA were treated for 48 weeks with certolizumab 200 mg every two weeks [71]. Patients who entered remission were randomized to continued therapy at 200 mg every two weeks, to dose reduction (ie, 200 mg every four weeks), or to placebo. After another 48 weeks, patients in both certolizumab groups were more likely to remain free of flare (80 versus 20.2 percent with placebo).

Remission achieved with low-dose TNF inhibition may be durable in some patients. In a two-year follow-up study of 106 patients with axial SpA in clinical remission who progressively reduced their dose of TNF inhibitors, tapering was successful in 52 percent [73]. However, only one patient was able to discontinue TNF inhibition entirely. Lower physician global score at baseline was one of the predictors of successful tapering.

●Discontinuing therapy – Some patients may tolerate discontinuation of therapy, although they may be at high risk for subsequent relapse. A systematic review and meta-analysis including five trials in patients with axial SpA showed that there is a high risk of a persistent flare after withdrawal of TNF inhibitors compared with continuing treatment [74]. In a subsequent study of 115 patients with nr-axSpA, only 25 percent of patients remained in remission for 40 weeks after stopping treatment with etanercept [75].

Interleukin 17 inhibitors

Overview — In general, anti-IL-17 antibodies are as effective as TNF inhibitors in controlling arthritis. They are probably more effective than TNF inhibitors in controlling psoriasis [76]. However, they are ineffective in suppressing bowel inflammation in patients with Crohn disease. They have also been reported to induce or exacerbate IBD [77]. Hence, they should be used with particular caution if needed for the treatment of SpA in patients with probable coexistence of IBD. They are also less effective in suppressing anterior uveitis [78].

There is limited information concerning the effect of anti-IL-17 antibodies on pregnancy and lactation. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Biologics with limited data'.)

Secukinumab — Secukinumab, an anti-IL-17A monoclonal antibody, is an alternative to TNF inhibitors for patients with axial SpA. Secukinumab can be initiated with or without a loading dose. Loading doses are 150 mg subcutaneously at weeks 0, 1, 2, 3, and 4, and then dosing is every 4 weeks. Without a loading dose, the dose is 150 mg subcutaneously every 4 weeks. For patients with continued high disease activity (eg, an ASDAS >2.1 or BASDAI >4.0), particularly those with a previous inadequate response to TNF inhibitors, the dose can be increased to 300 mg subcutaneously every 4 weeks [79-81]. Secukinumab can also be administered intravenously with or without loading dose. The loading dose is 6 mg/kg at week 0. The maintenance dose is 1.75 mg/kg every four weeks thereafter. Without a loading dose, the dose is 1.75 mg intravenously every four weeks. The maximum dose is 300 mg per infusion.

Patients should be screened for latent TB prior to use of this agent and should be treated appropriately, if screening positive, by initiation of anti-TB therapy before starting secukinumab. However, there are no reports, in contrast to TNF inhibitor therapy, of latent TB reactivation under secukinumab treatment. Hence, an IL-17 inhibitor, such as secukinumab (or ixekizumab (see 'Ixekizumab' below)), is the preferred biologic in patients in whom there is high risk for TB. (See "Tuberculosis infection (latent tuberculosis) in adults: Approach to diagnosis (screening)" and "Treatment of tuberculosis infection (latent tuberculosis) in nonpregnant adults without HIV infection".)

Secukinumab became commercially available in the United States in 2016 for use in AS, and in 2020 for nr-axSpA, irrespective of prior use of TNF inhibitors. This drug is also available for use in active AS and nr-axSpA in Europe and other parts of the world. Because of its more recent release and more limited number of indications, there is less experience with this agent than with the TNF inhibitors as a class and with the individual TNF inhibitors; it is also available for use in psoriasis and psoriatic arthritis.

The efficacy and relative safety of secukinumab has been demonstrated in four randomized trials in AS that compared the efficacy of secukinumab in several dosing regimens (75 mg, 150 mg, and 300 mg subcutaneously every four weeks following weekly doses for the first four weeks of therapy) with placebo [82-84]. In addition, secukinumab was beneficial, compared with placebo, both in patients who were naïve to TNF inhibitor therapy and those with prior TNF inhibitor exposure [82,83,85,86]. As an example, in one trial, these differences were 68 versus 31 and 50 versus 24 percent, respectively [86]. Responses were sustained after two to three years [87-89].

The efficacy and safety of secukinumab 150 mg every four weeks has also been demonstrated in a randomized trial in patients with nr-axSpA [90]. In total, 555 patients were randomly assigned to receive secukinumab 150 mg every four weeks with or without a loading dose, or to placebo. The proportion achieving an ASAS40 among TNF inhibitor-naïve patients was greater for the loading dose group (41.5 percent) at week 16 and non-loading dose group (39.8 percent) at week 52 compared with placebo (29.2 percent at week 16 and 19.9 percent at week 52) [90]. A long-term favorable safety profile is supported by four to five years of postmarketing surveillance data [91].

Infections, including candidiasis, were more common in patients receiving secukinumab during the placebo-controlled period of the first 16 weeks of the trials [84]; during the entire period of treatment, the rates of neutropenia, candida infection, and Crohn disease were each just below one case per 100 patient-years among the secukinumab-treated patients.

Ixekizumab — Ixekizumab, like secukinumab, is another anti-IL-17A monoclonal antibody available in the European Union and also the United States, where it has been approved by the US Food and Drug Administration (FDA) for the treatment of active AS and nr-axSpA. In AS, it is administered as a subcutaneous injection of 160 mg once, for the initial dose, followed by 80 mg every four weeks. In nr-axSpA, the initial dose is 80 mg. Subsequent doses are 80 mg every four weeks. It may be administered alone or in combination with conventional DMARDs (eg, SSZ). Similar to secukinumab, patients should be tested for latent TB prior to starting therapy, and patients should be monitored for symptoms of IBD, which has been described as a complication during the treatment.

Ixekizumab has been shown in randomized phase 3 trials to be effective in patients with active AS and active nr-axSpA [92-94].

●In a trial involving 341 patients with active AS/radiographic axial SpA with inadequate responses or intolerance to NSAIDs who had not previously received a biologic DMARD, patients randomly assigned to receive ixekizumab (80 mg subcutaneously every two or four weeks) or adalimumab (40 mg subcutaneously every two weeks) were more likely at week 16 to achieve an ASAS40 composite response measure compared with those receiving placebo (52, 48, and 36 versus 18 percent; differences of 33, 30, and 17 percent; 95% CI of 20-47, 16-43, and 4-30, respectively) [93].

●In a trial involving 316 patients with active AS/radiographic axial SpA with an inadequate response or intolerance to one or two TNF inhibitors, patients randomly assigned to receive ixekizumab (80 mg subcutaneously every two or four weeks) were more likely at week 16 to achieve an ASAS40 compared with those receiving placebo (31 and 25 versus 13 percent, p = 0.003 and 0.017, respectively) [92].

●In a trial involving 303 patients with active nr-axSpA with inadequate responses or intolerance to NSAIDs, patients were randomly assigned to receive ixekizumab (80 mg every two or four weeks) or placebo. Patients receiving ixekizumab were more likely to achieve an ASAS40 response (40 and 35 versus 19 percent, respectively) [94].

Ixekizumab can be safely tapered in patients who respond to initial therapy with this drug. In a trial involving 155 patients with axial SpA who achieved remission following 24 weeks of treatment with ixekizumab, patients were randomized to receive treatment with ixekizumab every two weeks, ixekizumab every four weeks, or placebo. Eighty-three percent of patients in both ixekizumab groups remained free of flare versus 55 percent of patients in the placebo group [95].

Another trial found that patients with axial SpA who achieved remission with ixekizumab and were continued on treatment were less likely to flare and had delayed time-to-flare when compared with patients who withdrew to placebo [95].

In all trials, treatment-emergent adverse effects, largely minor to moderate infections, were similar in frequency among the actively treated groups and more common than with placebo [84]. Injection site reactions were seen in some patients as well. The safety profile appears to be unchanged based on data from up to five years of follow-up [96,97].

A systematic review of the randomized trials comparing IL-17 inhibitors with placebo showed similar responses and safety in AS patients treated with secukinumab and ixekizumab, although no studies directly compared the two agents [98].

INADEQUATE RESPONSE OR INTOLERANCE TO INITIAL BIOLOGIC

Definitions — One should distinguish between two categories as the reason for switching agents [45]:

●Primary failure – A patient can be considered as having an inadequate response if there is no improvement (eg, an Ankylosing Spondylitis Disease Activity Score [ASDAS] decrease of >1.1) following 12 weeks of therapy with a given biologic agent. The Assessment of SpondyloArthritis International Society (ASAS) calculator (https://www.asas-group.org/instruments/asdas-calculator/) is accessible on the ASAS website and can be downloaded as an app. (See 'Evaluation and monitoring' below.)

●Secondary failure – Patients in whom there is initial improvement with drug therapy, but who experience a subsequent relapse (ie, worsening or recurrence of disease activity) are described as experiencing secondary failure for loss of efficacy. Patients who do not tolerate the particular tumor necrosis factor (TNF) or interleukin (IL) 17 inhibitor (eg, due to an adverse event) are defined as experiencing a form of secondary drug failure as well.

Primary and secondary failures are characterized by continuing symptoms of pain and/or stiffness that affect routine activities [1,99]. (See 'Evaluation and monitoring' below.)

Failure to respond to multiple biologic agents should prompt a re-evaluation of the initial diagnosis of axial spondyloarthritis (SpA), particularly if the diagnosis relied primarily on imaging findings. Primary failure can also be caused by nonadherence or comorbid conditions that may contribute to the patient's symptoms, such as fibromyalgia. Secondary failure should prompt evaluation for causes other than inflammation. Spinal fracture should be considered if compatible with clinical presentation [27]. (See 'Evaluation and monitoring' below.)

Approach to switching biologic agents — The choice of therapy in patients with inadequate efficacy or intolerance of a first or subsequent biologic is based in part upon the reason for drug discontinuation; it is also substantially influenced by an assessment of individual factors in a process of active shared decision-making involving the patient and their clinician, which is also affected by factors such as patient preferences for route and frequency of administration, the duration and severity of disease, and coexisting diseases or comorbidities, as well as regulatory and patient cost concerns [100].

There are three major alternatives for those patients who are unable to continue with the initial TNF inhibitor either because of side effects or ineffectiveness:

●Switch to another TNF inhibitor (see 'TNF inhibitors' above)

●Switch to an anti-IL-17 antibody (see 'Interleukin 17 inhibitors' above)

●Switch to a Janus kinase (JAK) inhibitor (see 'Janus kinase inhibitors' below)

We take the following general approach [45]:

●Primary failure to an initial TNF inhibitor – In patients with an inadequate response to a TNF inhibitor (typically after three months of therapy), we suggest an anti-IL-17 antibody or a JAK inhibitor, rather than different TNF inhibitor. Anti-IL-17 antibodies and JAK inhibitors have the theoretical appeal that the mechanisms of action are different from TNF inhibitors.

●Secondary failure to an initial TNF inhibitor – In patients with secondary drug failure to an initial TNF inhibitor from loss of efficacy to the initial agent, a reasonable option is a second TNF inhibitor, given the evidence of benefit, at least initially, with this class of drug.

●Intolerance to an initial TNF inhibitor – In patients who develop intolerance to the initial agent, especially if the intolerance is considered to belong to the TNF inhibitor class (eg, rash, injection site reactions, infection), switching to an anti-IL-17 antibody or a JAK inhibitor is a reasonable option.

●Secondary failure of two consecutive TNF inhibitors – Secondary failure of two consecutive TNF inhibitors does not preclude a response to a third TNF inhibitor or to an anti-IL-17 antibody or a JAK inhibitor. All three are reasonable options, although the likelihood of responding to a third TNF inhibitor is lower after two failures.

●Primary drug failures with TNF inhibitors, anti-IL-17 antibodies, and JAK inhibitors – In patients who have had an inadequate response to at least one TNF inhibitor, an anti-IL-17 antibody, and a JAK inhibitor, we repeat the diagnostic evaluation to ensure that the diagnosis is correct before using alternative therapies, with particular attention to concomitant fibromyalgia and to findings that suggest possible psoriatic arthritis [27]. (See "Clinical manifestations and diagnosis of fibromyalgia in adults" and "Clinical manifestations and diagnosis of psoriatic arthritis".)

It is especially important to exclude fibromyalgia as the cause of continued symptoms, as fibromyalgia can coexist with axial SpA with a prevalence of approximately 16.4 percent [32].

We also look for findings consistent with previously undiagnosed psoriasis involving skin or nails, which may suggest a diagnosis of psoriatic arthritis with axial SpA; in patients in whom a diagnosis of psoriatic SpA is likely, trials of agents that have proven effective in psoriatic arthritis, but that are typically not effective (or of unproven efficacy) in ankylosing spondylitis (AS) and nonradiographic axial SpA (nr-axSpA), may be reasonable options. Such agents include anti-IL-23 monoclonal antibodies [101]. (See "Diagnosis and differential diagnosis of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults" and "Treatment of psoriatic arthritis" and 'Resistant to standard therapies' below.)

Another option is to switch to another TNF inhibitor of different mode of action. With TNF inhibitors, etanercept is a soluble TNF receptor, the others are antibodies against TNF alpha.

●Inadequate response or intolerance of secukinumab or ixekizumab as initial agent – In the patients who may receive secukinumab or ixekizumab as the initial drug, but need to switch, we would use a TNF inhibitor as the next agent. In those unable to use a TNF inhibitor, we would use a JAK inhibitor next in such patients.

Efficacy of switching TNF inhibitors — In general, the frequency of good responses to TNF inhibitors is still substantial but decreases with the use of subsequent agents, compared with the first. This is well studied in patients with rheumatoid arthritis (see "Treatment of rheumatoid arthritis in adults resistant to initial biologic DMARD therapy") but has also been examined in patients with AS. There is little evidence and a lack of randomized trials to serve as a guide in managing these patients. Most reports are based on registry studies [45].

As an example, a 12-week, open-label study involving 1250 patients with AS evaluated the effectiveness of switching to adalimumab when patients failed etanercept or infliximab because of either an inadequate response or an adverse effect [102]. Such patients were compared with those begun on adalimumab who had never received etanercept or infliximab. After 12 weeks, a 50 percent improvement in the BASDAI response (BASDAI 50) from baseline was observed more often in patients who had never received TNF antagonists, compared with those who had received a prior TNF inhibitor (63 versus 41 percent, odds ratio [OR] 0.40, 95% CI 0.31-0.53).

Other studies involving patients with AS have also shown levels of benefit with second agents that approach those with the first drug. In one study, efficacy of etanercept in 23 patients with AS who were switched from infliximab due to loss of benefit or intolerance was studied at 54 weeks by the Assessment of SpondyloArthritis International Society [ASAS] 20 percent, 50 percent, and 70 percent improvement scores. The percent of patients reaching those scores was 74, 61 and 39, respectively [103]. As another example, benefits of a subsequent TNF inhibitor were described in a study of 47 patients with SpA (19 of whom had AS) that found similar good outcomes in patients who were switched from either infliximab or etanercept to adalimumab as a second or third agent, regardless of the reason for discontinuation [104].

One registry-based analysis in patients with an inadequate response to an initial TNF inhibitor found a subsequent TNF inhibitor to have comparable efficacy to switching to secukinumab [105].

Janus kinase inhibitors

Overview — JAK inhibitors, classified as targeted synthetic DMARDs, are oral synthetic drugs targeting the JAK enzymes. Depending upon the specific form of axial SpA, there are two agents potentially available for use in axial SpA; these are upadacitinib (see 'Upadacitinib' below) and tofacitinib (see 'Tofacitinib' below). Because of concern regarding adverse event risk, these agents are typically considered as one of the therapeutic options after an inadequate response to or intolerance of initial TNF inhibitor therapy.

JAKs initiate signal transduction pathways of a number of cytokines. The inhibitors have demonstrated usefulness in a number of inflammatory rheumatic diseases. The biology, principles of use, precautions (including screening for latent tuberculosis [TB]), and adverse effects of JAK inhibitors for rheumatologic diseases are described in detail separately. (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders".)

JAK inhibitors have two advantages. The first is that they are taken orally. The second is that they can be stored at room temperature. By contrast, all TNF inhibitors and anti-IL-17 antibodies need to be administered by injections and stored by refrigeration. JAK inhibitors are also useful for patients with psoriatic arthritis and for those with ulcerative colitis.

Nevertheless, we suggest that they should be used only after patients have demonstrated an inadequate response to one TNF inhibitor, given observations regarding increased cardiovascular, malignancy, and thrombotic risk of tofacitinib compared with TNF inhibitors in patients >50 with increased cardiovascular risk and resultant cautions issued by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA). (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders", section on 'Adverse effects'.)

Tofacitinib has been associated with higher risk of venous arterial thromboembolic events [106]. In addition, the FDA has issued a black box warning for JAK inhibitors as a class that there might be higher risks of cardiovascular events such as heart attacks and strokes, cancer, blood clots, and death. This warning is based on post-marketing analysis and highlights individuals aged >50 years [107,108]. The EMA safety committee stated that patients aged >65 years should be treated with tofacitinib only when no alternative treatment is available [109]. For upadacitinib, less safety information is available, which comes only from trials that included selected patients [110]. Patients should be screened for those risks prior to considering JAK inhibitors. Older age and smoking would be adverse risk factors. (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders", section on 'Pretreatment screening and precautions'.)

Patients taking JAK inhibitors have also been found to have a twofold increase in herpes zoster infection [110]. Recommendations regarding immunization against herpes zoster and other infectious agents for patients with axial SpA and other autoimmune and inflammatory rheumatic diseases are described in detail separately. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults".)

JAK inhibitors are contraindicated during pregnancy. (See "Safety of rheumatic disease medication use during pregnancy and lactation", section on 'Small molecules with limited data'.)

Upadacitinib — Upadacitinib has been approved by the European Commission for treatment of axial SpA (nr-axSpA and AS) and for psoriatic arthritis [111]. In the United States, it has been approved by the FDA for treatment of adults with either axial SpA (nr-axSpA and AS) or psoriatic arthritis who have had an inadequate response to or intolerance of one or more TNF inhibitors [112]. (See "Treatment of rheumatoid arthritis in adults resistant to initial biologic DMARD therapy", section on 'Upadacitinib' and "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Upadacitinib'.)

The benefits of 15 mg once daily of upadacitinib in AS were first shown in a randomized phase 2/3 trial involving 187 patients with active AS and inadequate responses or intolerance to nonsteroidal antiinflammatory drug (NSAID) therapy, in which patients treated with upadacitinib, compared with placebo, were more likely to achieve an ASAS40 response at 14 weeks (52 versus 26 percent) [113]. On extending this study to 64 weeks, partial remission was observed in approximately 35 percent of patients on upadacitinib [114].

Upadacitinib appears to be effective for patients who have had an inadequate response to biologic (b)DMARDs. In a trial including 420 patients with active AS refractory to one or two of either TNF inhibitors or IL-17 inhibitors, more patients in the upadacitinib group compared with placebo group achieved an ASAS40 response at 14 weeks (45 versus 18 percent) [115].

Upadacitinib is also effective for patients with nr-axSpA. A trial of 313 patients with active nr-axSpA detected a higher ASAS40 response rate among patients administered upadacitinib compared with placebo at 14 weeks (45 versus 23 percent) [116].

The major potential side effects of upadacitinib and other JAK inhibitors include venous thrombosis, pulmonary embolism, infections including TB, and malignancy [117]. (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders", section on 'Adverse effects'.)

Tofacitinib — Tofacitinib is a JAK inhibitor that became available in the United States at the end of 2021 for use in patients with active AS who have had an inadequate response or intolerance to at least one or more TNF inhibitors [118]. It was already available in the United States for use in psoriatic arthritis, rheumatoid arthritis, and ulcerative colitis. In a phase 3 randomized trial of 269 patients with AS comparing tofacitinib 5 mg twice daily with placebo, improvement was observed as early as 2 to 4 weeks after starting therapy. An ASAS40 response at week 16 was more likely with tofacitinib (40.6 versus 12.5 percent). The response was sustained at 48 weeks [119].

RESISTANT TO STANDARD THERAPIES

Lack of role for systemic glucocorticoids for axial disease — Systemic glucocorticoids are not indicated for patients with axial spondyloarthritis (SpA). Very limited data suggest that only very high doses of prednisolone may have some benefit for very short-term therapy [120].

Agents of possible benefit — Little evidence and no randomized trials are available to determine the best treatment options for patients in whom two to three different tumor necrosis factor (TNF) inhibitors and interleukin (IL) 17 inhibitors have all been inadequate. A third (but investigational) IL-17 inhibitor, bimekizumab, has been reported to be effective in patients with active ankylosing spondylitis (AS) [121]. Additional drugs have been reported in limited studies, but typically without confirmation, to be somewhat effective in SpA.

Used in other disorders

●Thalidomide – The potential benefit of thalidomide has been examined in AS because of its immunomodulatory properties, including its effects on TNF [122-124]. A 2002 review of observational studies reported that among more than 50 patients treated with thalidomide, 68 percent improved and 19 percent withdrew from treatment due to lack of efficacy or adverse effects [124]. Thalidomide has also been reported to reduce relapses after the discontinuation of etanercept [125]. However, the seriousness of the potential side effects will need to be considered in determining whether it has a role in therapy, and randomized trials have not been reported. It is rarely, if ever, used in the United States or Europe for AS.

●Brodalumab – Brodalumab is an anti-IL-17 receptor-A monoclonal antibody that has been approved for use in plaque psoriasis. A phase 3 study of 159 axial SpA patients in Asia (Japan, Korea, and Taiwan), which compared brodalumab with placebo, found a greater frequency of an ASAS40 response at week 16 with brodalumab (approximately 40 versus 20 percent) [126]. (See "Treatment of psoriatic arthritis".)

●Bimekizumab – Bimekizumab is a monoclonal antibody that neutralizes IL-17A and IL-17F. A phase 2b study of 303 patients with AS that compared bimekizumab with placebo found that patients were more likely to achieve an ASAS40 response at week 12 with the active agent (46.7 versus 13.3 percent) [121]. An open-label extension of this study indicates that the efficacy of bimekizumab is sustained after 3 years of treatment (ASAS40 response 53.7 percent) [127].

Biologic and other agents effective in other diseases but not axial SpA — Several agents that have been shown to be effective in patients with psoriatic arthritis, rheumatoid arthritis, or other forms of inflammatory arthritis have been found to be ineffective in patients with axial SpA or insufficiently effective to warrant further development for this condition, including:

●Abatacept, a T-cell costimulation blocker [128]

●Tocilizumab and sarilumab, IL-6 inhibitors [129,130]

●Ustekinumab, an anti-IL-12/23 p40 monoclonal antibody [131]

●Risankizumab, an anti-IL-23 p19 monoclonal antibody [132]

●Apremilast, an oral phosphodiesterase 4 (PDE4) inhibitor [133,134]

●Rituximab, a monoclonal anti-CD20 antibody that depletes B cells [135]

Role of non-NSAID analgesics — In patients who need additional temporary pain relief, not adequately provided by NSAIDs, we most often use acetaminophen (500 to 1000 mg up to three times daily), which can be added to NSAIDs if required temporarily [136].

Opioids should generally be avoided and used, if needed, for only a short period of time. As an example, tramadol (25 to 50 mg up to four times daily) has sometimes been used if opioids are required, but other opioid medications should be avoided in axial SpA, and such situations are rare.

There are few studies describing the usefulness of analgesics or opioids in axial SpA, whether alone or in combination with other drugs [137].

Isolated active sacroiliitis — A treatment option used by one of the authors (DY) in patients refractory to NSAIDs with severe isolated sacroiliitis is injection of long-acting glucocorticoids (eg, 40 mg triamcinolone) into the sacroiliac joints. Small studies with methodologic limitations suggest this approach may be beneficial in patients who complain of marked pain at the sacroiliac joints that is unresponsive to systemic medications [2,138]. It is preferable to carry out the injections under imaging (usually fluoroscopic) guidance [139].

Benefit of injection of long-acting glucocorticoids has been found in some but not all studies [140,141]. However, one randomized trial reported more than 70 percent relief in over 80 percent of injected sacroiliac joints [141]. Relief persisted for as long as six months or more after the injection and no complications were seen.

EVALUATION AND MONITORING — The response to therapies as well as side effects of agents vary tremendously from patient to patient. Each patient should be followed regularly for monitoring of disease activity and medication safety. The frequency of visits and laboratory assessment depend upon the patient's response to therapy and the risks of the medications being used. Patients initially require evaluation every two to four weeks to assess the response to nonsteroidal antiinflammatory drugs (NSAIDs), then monthly to every three months once stable or when starting a biologic agent. Patients should be evaluated after at least 12 weeks of tumor necrosis factor (TNF) inhibitor or anti-interleukin (IL) 17 therapy. Subsequent monitoring can take place every three to six months when the disease is under control.

Imaging should be repeated only when the results might alter the treatment strategy [4]. For patients with axial spondyloarthritis (SpA), it is not necessary to repeat magnetic resonance imaging (MRI) of the spine or the pelvis to confirm that there is a response to therapy, nor is it necessary to repeat radiographs of the spine or sacrum at a fixed scheduled interval.

Clinical assessment should include a focused history and examination directed at the patient's known manifestations and screening for other features associated with axial SpA (see "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults"). The adequacy of the response is based upon a combination of measures:

●Disease activity assessment with either the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) or the Ankylosing Spondylitis Disease Activity Score (ASDAS):

•BASDAI – The BASDAI is an instrument for the assessment of disease activity that is presented in a questionnaire format. A BASDAI score of ≥4 (on a scale of 0 to 10) is indicative of active disease that warrants consideration of biologic therapy (table 2). Clinically significant improvement is defined as either a 50 percent improvement of the BASDAI score (BASDAI 50) or an absolute change of ≥2 on a scale of 0 to 10 and a clinical "expert" opinion that a particular patient has improved.

•ASDAS – The ASDAS is a composite instrument to measure disease activity. Unlike the BASDAI, the ASDAS also incorporates information from the level of acute phase reactants in addition to clinical parameters (a patient global score and several questions also used for the BASDAI). An ASDAS calculator is accessible at the Assessment of SpondyloArthritis International Society (ASAS) website (which can also be downloaded on mobile devices through an ASAS app, available in several languages), together with information regarding its use and interpretation. The ASDAS categorizes the disease activity as inactive, low, high, or very high [142]. The disease activity should at least be high (≥2.1) to warrant consideration of biologic therapy. A change of ≥1.1 in the ASDAS score is considered a significant improvement, while a change of ≥2.0 is a major improvement. A flare of disease activity has been defined as an increase of the ASDAS by ≥0.9 points [143].

●Individualized contextual judgment – Over the course of therapy and at key decision points, each patient's personal goals and preferences should be incorporated in treatment decisions, particularly regarding the targets of treatment, and patients should be assessed individually in the context of their own psychological, social, and clinical history; physical examination; previous therapies; laboratory tests; results of imaging; and degree of response to prior therapies [3,100]. Use of a serial quantitative measure is of particular benefit; scoring the patient global assessment on a scale of 0 to 10, which is included in the ASDAS calculation, should be performed even if an ASDAS or BASDAI is not.

PERIPHERAL ARTHRITIS AND PERIARTICULAR DISEASE — The approach to patients with predominantly peripheral spondyloarthritis (SpA) differs from that for patients with symptomatic axial SpA, whether or not peripheral SpA is also present in such patients, because of the potential benefit of traditional (conventional synthetic [cs]) disease-modifying antirheumatic drugs (DMARDs) for peripheral joint manifestations but not for the axial disease. By contrast, treatments for the axial disease, including biologic agents, are typically also quite effective for the peripheral manifestations. (See "Treatment of peripheral spondyloarthritis", section on 'Resistant to nonbiologic DMARD'.)

SURGERY — Hip and spine surgery may be beneficial in selected patients with axial spondyloarthritis (SpA). Clinicians and anesthesiologists should be cautioned about perioperative risks such as reduced chest expansion and more rigid cervical spines. Indications for surgery are:

●Severe hip involvement, with persistent pain or severe limitation in mobility and quality of life

●Atlantoaxial subluxation with neurologic impairment

●Severe flexion deformities with impaired ability to look in a forward direction

Total hip arthroplasty — Total hip arthroplasty (THA, total hip replacement) is indicated whenever there is severe, persistent pain or severe limitation in mobility and quality of life due to hip involvement. THA is more common in patients with early onset of disease, axial, and entheseal involvement [144]. (See "Total hip arthroplasty".)

Almost 90 percent of patients will experience pain relief from the hip arthroplasty, with improved ranges of motion, 90 percent survival of the replaced hip for 10 years, and 72 percent for 15 years [145].

Patients with ankylosing spondylitis (AS) may be at higher risk of developing heterotopic ossification following joint replacement, but this remains an unusual complication [146]. If, however, this complication has occurred following a prior joint arthroplasty, prophylactic therapy has been recommended, such as a nonsteroidal antiinflammatory drug (NSAID) beginning on the day of surgery or radiation therapy (pre- or postoperative) [145]. (See "Complications of total hip arthroplasty", section on 'Heterotopic ossification'.)

Spinal surgery — There are several indications for spinal surgery. The first is acute fracture. The most common spinal fracture is in the cervical spine. Approximately 25 percent of cervical fractures are associated with spinal cord injuries [147].

Cervical fusion is indicated for the very small number of patients who develop atlantoaxial subluxation with impairment in neurologic function. This problem is managed in a fashion similar to that in rheumatoid arthritis. (See "Cervical subluxation in rheumatoid arthritis".)

Corrective wedge osteotomy is an effective surgical treatment for those with kyphosis with functional impairment [148].

SPECIAL POPULATIONS

Patients with axial spondyloarthritis and extraarticular manifestations — Extraarticular manifestations such as uveitis, psoriasis, and inflammatory bowel disease (IBD) frequently occur in patients with axial spondyloarthritis (SpA). The presence of an extraarticular manifestation may add complexity to care and influence treatment decisions, which should be coordinated with the other clinicians caring for the patient.

●Uveitis – Unilateral anterior uveitis is the most common extraarticular complication of axial SpA, occurring in 25 to 40 percent of patients [149]. The evaluation and management of uveitis are discussed in detail separately (see "Uveitis: Etiology, clinical manifestations, and diagnosis" and "Uveitis: Treatment", section on 'Noninfectious uveitis'). Use of anti-tumor necrosis factor (TNF) therapy may also decrease the frequency of recurrences of uveitis in patients with axial SpA [78,150,151].

●Psoriasis – Psoriasis is present in up to approximately 10 percent of patients with axial SpA [149]. Patients with concomitant psoriasis have more frequent peripheral joint involvement and possibly a more severe axial SpA disease course compared with axial SpA patients without psoriasis [152,153]. Peripheral arthritis is usually treated initially with conventional synthetic (cs) disease-modifying antirheumatic drugs (DMARDs) such as sulfasalazine (SSZ) or methotrexate (MTX). Patients resistant to DMARDs or with axial disease that failed treatment on two nonsteroidal antiinflammatory drugs (NSAIDs) can be treated with a biologic. The presence of psoriasis can influence the choice of biologic agents. There are agents that are effective in patients with psoriatic arthritis but not in those who have axial SpA without psoriasis. (See "Treatment of psoriatic arthritis".)

●Inflammatory bowel disease – Overt IBD occurs in approximately 5 to 10 percent of patients with axial SpA, and approximately 4 to 10 percent of patients with IBD have concomitant findings of axial SpA [149,154]. The presence of IBD impacts the management approach. Patients with IBD should avoid anti-interleukin (IL) 17 antibodies. The clinical manifestations, diagnosis, and treatment of peripheral and axial SpA associated with IBD are described in detail separately. (See "Clinical manifestations and diagnosis of arthritis associated with inflammatory bowel disease and other gastrointestinal diseases" and "Treatment of arthritis associated with inflammatory bowel disease".)

Pregnancy — Hormonal status and fertility are normal. The effect of pregnancy on disease activity is variable [155]. Considerations relevant to the use of medications for axial SpA during pregnancy and lactation are reviewed separately. In general, both anti-IL-17 antibodies and Janus kinase (JAK) inhibitors should be avoided. (See "Safety of rheumatic disease medication use during pregnancy and lactation".)

Axial spondyloarthritis and osteoporosis — Low bone mineral density (BMD) and bone loss have been documented in patients with axial SpA and are evident within the first 10 years of disease [156-158]. Screening and treatment of osteoporosis are discussed in detail elsewhere. (See "Screening for osteoporosis in postmenopausal women and men" and "Evaluation and treatment of premenopausal osteoporosis" and "Overview of the management of low bone mass and osteoporosis in postmenopausal women" and "Treatment of osteoporosis in men".)

PROGNOSIS — Axial spondyloarthritis (SpA) is a chronic disease, although a minority of patients may experience a spontaneous remission. Most axial SpA patients with mild disease that is restricted to a small area of involvement are able to maintain almost full functional and employment capacity. The most serious musculoskeletal consequences are unremitting spinal pain, hip destruction, and spinal fusion. Only a minority of patients develop life-threatening extramusculoskeletal complications.

A questionnaire survey done prior to the availability of the biologic agents found that employment was affected in approximately 30 percent of the males with ankylosing spondylitis (AS) [159]. However, the majority of patients were able to work well into their fifties. Without biologic therapies, disease activity usually fluctuated in the individual patient, with symptoms usually persisting over decades, and approximately 1 percent of patients developed a stage of "burnt-out" disease activity and entered long-term remission [160].

The prognosis for axial SpA may have improved since the 1990s. This may relate to the availability of tumor necrosis factor (TNF) inhibitors beginning at the end of that decade, but the evidence is indirect and does not prove causality. In a longitudinal study of 350 patients with AS, approximately a quarter of the patients showed radiographic progression of spinal disease over two years [161]. The rate of radiographic progression was reduced by 20 percent in patients using TNF inhibitors. TNF inhibitors might be more effective in controlling inflammation when used early in the disease course, prior to the development of changes on plain radiographs [49].

Effects of treatment on radiographic progression — Whether long-term use of anti-TNF agents can halt the radiographic progression of AS has long been uncertain because such effects may not be evident until patients have been followed for more than two years after starting the medication [162-165], and the arrest of progression might not start appearing until more than two years after starting the TNF inhibitors [166]. A 2020 structured literature review suggested that TNF inhibitors were able to slow down radiographic progression of disease in the spine [167]. Some examples of the evidence include:

●Data from 432 AS patients from the Swiss Clinical Quality Management cohort analyzed the impact of TNF inhibitors on spinal radiographic progression [168]. This study could demonstrate that TNF inhibitors are associated with a reduction of spinal radiographic progression in patients with AS, but this effect is mediated through the inhibiting effect of TNF inhibitors on disease activity measured by the Ankylosing Spondylitis Disease Activity Score (ASDAS). The odds ratio (OR) of radiographic progression halved in patients who have used TNF inhibitors for more than two years (OR 0.50, 95% CI 0.28-0.88). In patients who achieved an inactive disease status (ASDAS ≤1.3), radiographic progression was entirely inhibited. Longer duration of TNF inhibition was associated with more reduction in radiographic progression (≥4 years of treatment showed a 70 percent lower estimate in radiographic progression; in ≤4 years of treatment, this was 45 percent) [168].

●In a prospectively followed cohort of 334 patients treated with standard therapies for AS, in whom TNF inhibitors were administered to 201 patients, treatment with a TNF inhibitor was associated with a 50 percent reduction in the odds of radiographic progression (OR 0.52, 95% CI 0.30-0.88) [169]. Such benefit was more evident in patients in whom these agents were begun earlier in the disease course and in whom the follow-up was at least four years.

●In an 18-year real-world study in Korea, the analysis was limited to patients who were initiated on TNF inhibitors, but compared radiographic progression during the follow-up intervals in which patients were on TNF inhibitors with intervals when they were not [170]. Radiographic assessments were performed at average intervals of 2.4 years. Radiographic progression was lower during the on-TNF inhibitors intervals compared with the off-TNF inhibitors intervals, consistent with a therapeutic effect. Additionally, radiographic progression was less severe in women than men, and more severe in those with associated eye involvement.

There is limited information on radiographic progression for patients with nonradiographic axial SpA (nr-axSpA) [167].

Some evidence also suggests that biologic therapy may prevent or reduce the need for joint arthroplasty. As examples:

●An analysis of data from the national arthroplasty registry in Norway compared surgical rates from 1988 to 2002 with those from 2003 to 2010 [171]. It described a trend since 2002 for a reduced frequency of hip replacement surgery in patients with AS compared with a continued increase in hip replacements in patients with osteoarthritis (OA) during this same time period. The onset of the apparent decline in need for joint replacement coincided with the introduction of TNF inhibitors in Norway between 2000 and 2003.

●Similarly, a study from the United States compared rates of hip, knee, and shoulder arthroplasties from 1991 to 2005 using data from large administrative databases in New York and California; the rate of arthroplasties for noninflammatory conditions increased by nearly 200 percent from 1991 to 2005, compared with an increase in rates of arthroplasty for SpA of only approximately 50 percent during the same interval [172]. The mean age at the time of arthroplasty decreased for noninflammatory causes (age 71.5 versus 69.0) but increased for patients with SpA (age 54.3 versus 60.4), also consistent with a reduction in the rate of joint injury for SpA since the introduction of TNF inhibitors.

Prognostic indicators — A number of prognostic indicators were identified before the availability of biologics in patients with AS. One study, for example, evaluated 328 patients with SpA; seven variables at entry correlated with increased disease severity [173]:

●Hip arthritis – OR 23

●Sausage-like finger or toe – OR 8

●Poor efficacy of nonsteroidal antiinflammatory drugs (NSAIDs) – OR 8

●High erythrocyte sedimentation rate (ESR; >30 mm/h) – OR 7

●Limitation in range of motion of the lumbar spine – OR 7

●Oligoarthritis – OR 4

●Onset less than 16 years of age – OR 3

A mild outcome was likely if none of these factors was present at entry (sensitivity 93 percent, specificity 78 percent). However, a severe outcome was predictable if the hip was involved or if three factors were present (sensitivity 50 percent), and mild disease could be virtually excluded (specificity 98 percent).

The following parameters have been found to be independent baseline predictors of a good response to TNF inhibitor therapy after three to six months of therapy [174] (see 'TNF inhibitors' above):

●Increased acute phase reactants

●Higher disease activity

●Higher functional status

●Younger age

●Human leukocyte antigen (HLA)-B27 positivity

Other factors have also been associated with a poor outcome in patients with AS [175-177]. These include cigarette smoking, increasing severity of radiographic changes, active disease as assessed by a disease activity index, functional impairment as assessed by a self-report, lower educational attainment, presence of other diseases related to SpA (eg, psoriasis, inflammatory bowel disease [IBD]), male sex, a history of uveitis, and occupational activities involving either dynamic flexibility (ability to quickly and repeatedly bend, twist, and stretch) or exposure to whole-body vibration (such as driving a truck or operating heavy equipment). Other predictors of poor outcome are presence of HLA-B27, older age, the presence of enthesitis, poor functional ability, and elevated C-reactive protein (CRP) [178].

Elevated CRP levels are associated with increased risk of radiographic progression both in the sacroiliac joints and in the spine [179,180]. Other factors associated with an increased degree of radiographic progression include the baseline severity of radiographic change and cigarette smoking [169,181]. A 12-year prospective follow-up study involving the Outcome in Ankylosing Spondylitis International Study (OASIS) cohort found that progressive radiographic spinal changes occurred significantly faster in men, HLA-B27-positive patients, and those with greater radiographic change at baseline [182]. Although long-term radiographic progression varied between patients (ranging from some showing no progression at all to others progressing rapidly over short periods of time), it continued over decades and followed a roughly linear course for the cohort as a whole.

Complications

●Spinal cord injury – Patients with AS suffer an increased rate of spinal fractures [147]. In Finland, the incidence of spinal cord injury among those with AS has been estimated to be increased more than 10-fold compared with the general population [183]. In those with AS, a majority of spinal cord injuries resulted from slips and falls, an event that was rarely the cause of cord damage in those without AS (53 versus 7 percent, respectively).

●Cardiovascular risk – AS is associated with an increased risk of cardiovascular diseases, including disease of the aortic root and rarely of the aortic valve, acute coronary syndromes (ACS), strokes, venous thromboembolism, and conduction abnormalities. Although cardiovascular involvement has been examined in patients with AS in several studies [184-186] and evaluated in patients with SpA in general [187], it has not been examined specifically in patients with nr-axSpA. Most but not all studies show increases in ischemic heart disease compared with the general population. The increased risk has been attributed to the systemic inflammation and increased prevalence of traditional cardiovascular risk factors, but also the use of NSAIDs. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Cardiovascular disease'.)

●Lymphoma risk – Most population-based studies demonstrate that the risk of developing lymphoma is not significantly increased in patients with AS, in contrast to some other chronic rheumatic diseases (eg, rheumatoid arthritis and Sjögren's disease) [188]. However, patients with AS older than 65 years may have an increased risk of non-Hodgkin lymphoma (standardized incidence ratio [SIR] 1.39, 95% CI 1.17-1.92), chronic lymphocytic leukemia (SIR 1.50, 95% CI 1.17-1.92), and multiple myeloma (SIR 1.52, 95% CI 1.12-2.06) [189].

●Mortality – Overall, mortality may be modestly increased [190]. This was illustrated in a review of eight studies on mortality in AS that concluded that there was an increase in mortality rates compared with the general population (standardized mortality ratios ranging from 1.32 to 2.62) [191]. In a registry study of the total adult population of Norway, mortality among patients with AS was approximately 1.4 times higher than that of the general population (hazard ratio [HR] 1.38, 95% CI 1.28-1.38) [192]. The major causes of death in patients with AS were infections, cancer, and diseases of the respiratory and cardiovascular systems [193,194].

GUIDELINES OF MAJOR ORGANIZATIONS — Management recommendations for ankylosing spondylitis (AS) and non-radiographic axial spondyloarthritis (nr-axSpA) have been developed jointly by the Assessment of Spondylo-Arthritis international society (ASAS) and by the European Alliance of Associations for Rheumatology (EULAR; formerly known as European League Against Rheumatism) [27] and jointly by the American College of Rheumatology (ACR), the Spondylitis Association of America (SAA), and the Spondyloarthritis Research and Treatment Network (SPARTAN) [2,4]. Our approach is generally consistent with these guidelines. The ACR/SAA/SPARTAN 2019 update and supplement to the previous guideline document addresses the use of agents not discussed in their older publication (eg, secukinumab, ixekizumab, tofacitinib, and tumor necrosis factor [TNF] biosimilars) and cautions regarding discontinuation or tapering of biologic agents in patients with stable AS, among other recommendations [4].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Spondyloarthritis".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Ankylosing spondylitis (The Basics)")

●Beyond the Basics topic (see "Patient education: Axial spondyloarthritis, including ankylosing spondylitis (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Goals of therapy – The primary goals of management are to optimize short- and long-term health-related quality of life through relief of symptoms, maintenance of function, prevention of spinal complications, minimization of extraspinal and extraarticular manifestations and comorbidities, and maintenance of effective psychosocial functioning. Most patients benefit from care by an expert in rheumatologic disease, such as a rheumatologist; care should be coordinated with appropriate specialists, depending upon the clinical features; and active patient engagement in shared decision-making with their clinical team is an important element of care. (See 'Goals and general principles of management' above.)

●Patient education – All patients with axial spondyloarthritis (SpA) should receive education about their disease and its management, counseling regarding smoking cessation, depression screening and psychosocial support, and physical therapy with instruction in home exercises. (See 'Nonpharmacologic interventions' above.)

●Initial therapy – For initial therapy of most patients with axial SpA, we suggest nonsteroidal antiinflammatory drugs (NSAIDs) rather than a biologic agent (Grade 2C). The addition of disease modifying-antirheumatic drugs (DMARDs) is typically not indicated, and for some patients, NSAIDs are the only medications required. Examples of NSAIDs used for these patients include naproxen (up to 500 mg twice daily) and ibuprofen (up to 800 mg three times daily), although any NSAID may be effective (table 1). Regardless of the NSAID used, the maximum dose is often required, and the response should be assessed after a sustained dose on a daily basis for at least two to four weeks. In patients with an inadequate response, we switch to a second NSAID. (See 'Initial drug therapy with NSAIDs' above and 'Duration of NSAID therapy' above.)

●Inadequate response to NSAIDs – For patients with an inadequate response to initial therapy with at least two different NSAIDs consecutively, we suggest adding a tumor necrosis factor (TNF) inhibitor (Grade 2C). Any of the TNF inhibitors may be used, and they do not need to be used together with a conventional synthetic (cs) immunosuppressive agent such as methotrexate (MTX). A TNF inhibitor (except etanercept) is specifically preferred in patients with inflammatory bowel disease (IBD), given that IBD may be exacerbated by anti-interleukin (IL) 17 agents. (See 'Inadequate response to NSAIDs' above and 'TNF inhibitors' above.)

●Alternatives to TNF inhibition

•IL-17 inhibitors – The IL-17 inhibitors secukinumab and ixekizumab are reasonable alternatives to a TNF inhibitor as initial biologic therapy (for example, in case of concomitant psoriasis), although there is much more experience and evidence of long-term efficacy and safety with the TNF inhibitors. IL-17 inhibitors are preferred in patients with psoriasis. (See 'Interleukin 17 inhibitors' above.)

•JAK inhibitors – Because of concern regarding adverse event risk with Janus kinase (JAK) inhibitors, we typically consider them only after an inadequate response to or intolerance of initial TNF inhibitor therapy. TNF inhibitors, anti-IL-17 inhibitors, and JAK inhibitors have not been directly compared in patients with axial SpA; however, all have similar levels of efficacy compared with placebo. (See 'Janus kinase inhibitors' above.)

●Inadequate response to biologic agents (bDMARD) – The choice of subsequent therapy in patients with inadequate efficacy or intolerance of a first biologic agent is based in part upon the reason for drug discontinuation. We prefer the following approach (see 'Approach to switching biologic agents' above):

•For patients with an initial response to a TNF inhibitor followed by a loss of efficacy, we switch to a second TNF inhibitor. (See 'TNF inhibitors' above.)

•For patients with an inadequate response (typically after three months of therapy for each TNF inhibitor used) or intolerance to one or two TNF inhibitors, we switch to an anti-IL-17 antibody (secukinumab or ixekizumab) or a JAK inhibitor (tofacitinib or upadacitinib). (See 'Interleukin 17 inhibitors' above and 'Janus kinase inhibitors' above.)

•For patients who were initially treated with an IL-17 inhibitor and responded inadequately after three months, we switch to a TNF inhibitor. (See 'TNF inhibitors' above.)

•For patients who have not responded to biologic therapies and JAK inhibitors, we reassess the diagnosis. Particularly important is careful consideration of coexisting fibromyalgia. (See 'Resistant to standard therapies' above and 'Agents of possible benefit' above.)

●Tapering therapies – Patients who have achieved sustained remission with the use of a bDMARD (ie, TNF inhibitor or IL-17 inhibitor) may be able to taper gradually. Tapering bDMARDs should be considered only in patients who have achieved low disease activity for at least six months. It is unclear whether this strategy can be extended to patients who have achieved remission using JAK inhibitors. (See 'Duration and tapering of therapy with TNF inhibitors' above.)

●Surgical intervention – Total hip replacement or cervical fusion may be appropriate for patients with severe hip involvement or atlantoaxial subluxation that impairs the patient's quality of life. (See 'Surgery' above.)

●Limited role for opiates or glucocorticoids – In patients with axial SpA, there is an extremely limited role, if any, for the use of opioid analgesics. Systemic glucocorticoids (eg, prednisone) are ineffective for axial SpA in low to moderate doses and are not indicated. (See 'Role of non-NSAID analgesics' above and 'Lack of role for systemic glucocorticoids for axial disease' above.)

●Prognosis – Most axial SpA patients with mild disease are able to maintain almost full functional and employment capacity. The most serious musculoskeletal consequences are unremitting spinal pain, hip destruction, and spinal fusion. Patients with axial SpA are at increased risk of spinal cord injury and cardiovascular disease. Whether TNF inhibition prevents radiographic progression of disease is not clear. (See 'Prognosis' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David Yu, MD, who contributed to earlier versions of this topic review.