Abstract
Giant cell arteritis (GCA) is the most common form of systemic vasculitis. It is a potentially severe disease with 25% of patients suffering vision loss or stroke. Our treatment paradigm is based on glucocorticoids. Glucocorticoids are required in high doses for prolonged periods and subsequently are associated with a significant amount of treatment-related morbidity. Alternative treatment options are urgently needed to minimize these glucocorticoid adverse events. Many other agents, such as methotrexate and tumour necrosis factor alpha inhibitors have been used in GCA, with limited or no evidence of benefit. Our emerging understanding of the pathogenic processes involved in GCA has led to an increased interest in the use of biologic agents to treat the disease. Two randomized controlled trials have recently reported dramatic effects of the use of the interleukin-6 targeted biologic tocilizumab in GCA, with significant increases in remission rates and decreases in glucocorticoid burden. While encouraging, longer-term and additional outcomes are awaited to clarify the exact positioning of tocilizumab in the treatment approach. Emerging data for other biologic agents, particularly abatacept and ustekinumab, are also encouraging but less well advanced. We are at the dawn of a new era in GCA treatment, but uncertainties and opportunities abound.
Introduction
Giant cell arteritis (GCA) is the most common form of systemic vasculitis. GCA occurs in those over the age of 50 and becomes increasingly common with ageing. 1 GCA can affect any medium or large artery and has a particular predilection for the aorta and its extracranial branches. 2 While the most common features of GCA are headache and polymyalgia, GCA is a potentially devastating disease with up to 25% of patients suffering vision loss or stroke. 3 In the longer-term, aneurysmal disease resulting from vascular inflammation is a major cause of morbidity and mortality. 4 Aortic aneurysm is the most common of these, occurring in 18% of patients; aortic dissection occurs in 5% of patients and is associated with a significantly reduced median survival of 1.6 years.4,5 Pathologically, GCA is characterized by a granulomatous inflammatory infiltrate. 2 The pathogenesis remains to be fully elucidated with both the innate and adaptive immune systems appearing to play a role. 6 Emerging evidence implicates a dual T-lymphocyte pathway with important roles for TH1- and TH17-driven inflammatory cascades. 7 The fact that the pathogenesis is both complex and incompletely understood has played a role in the limitations of current treatment options. Our emerging understanding of the pathogenesis of GCA will hopefully enhance the treatment approach in this disease. Our treatment of GCA has been based to date almost entirely on glucocorticoids. Multiple agents have offered promise, but ultimately little or no evidence of clinical efficacy. High-quality recent evidence has emerged showing preliminary efficacy of a number of biologic agents in GCA. Physicians treating patients with GCA should be aware of these developments. Details of the search strategy used for this review are given in Box 1.
Search strategy for review
We searched PubMed for the terms ‘giant cell arteritis’ OR ‘temporal arteritis’ AND ‘biologic’ OR ‘treatment’. Reference lists of obtained studies were also screened for additional relevant articles.
Treatment of GCA
The treatment of GCA in terms of both chronology and the individual patient is characterized by early successes followed by later difficulties and complications. Glucocorticoids were introduced to the treatment paradigm nearly 70 years ago and remain the cornerstone of treatment. Multiple other agents have either had a less than anticipated benefit, been ineffective or even harmful.8–10 For the individual patient, dramatic early treatment responses are tempered by later treatment-related complications and relapses.11–13
Glucocorticoids
Glucocorticoids were first used in GCA by Horton in 1949. 14 Formal confirmation of their efficacy was demonstrated by Shick in the 1950s. 15 Glucocorticoids rapidly improve symptoms and normalize inflammatory markers in GCA and are recommended as the treatment of choice in GCA in consensus guidelines.16,17 Glucocorticoids are required in high doses for prolonged periods in order to induce and maintain remission in GCA; however, there is considerable uncertainty over the optimum dosing and tapering schedules. 18 Most guidelines recommend commencing on a dose of 40–60 mg daily with a subsequent individualized tapering regime adjusted if necessary by symptoms and adverse events.16,17 This tapering regime is adjusted on an individual basis and may require temporary increases in steroid doses in the event of relapses. As is evident from the doses illustrated in this regime, patients with GCA receive a high median cumulative steroid dose, equivalent to 6.5 g of prednisolone. 12 Glucocorticoids incompletely extinguish the vascular inflammation in GCA as evinced by findings in animal and autopsy studies.19,20 A study of repeat Temporal artery biopsies (TABs) in patients treated with high-dose glucocorticoids for GCA showed evidence of persistent vascular inflammation in 75% of patients at 6 months and 44% at 12 months, despite clinically well-controlled disease activity. 21 The consequences of this persistent vascular inflammation include risks of disease relapse and the development of long-term complications such as aortic aneurysms.5,22 Glucocorticoids are associated with significant treatment-related complications in up to 95% of patients.11,12 These adverse events are frequently serious and even life-threatening developments such as fractures, diabetes mellitus, hypertension and sepsis.11,12 Glucocorticoids remain indispensable in the treatment of GCA and provide a rapid and effective treatment; however, they are associated with short-term and long-term complications and patients continue to experience overt relapses and subclinical disease progression. Therefore, other treatment modalities are urgently needed.
Methotrexate
Methotrexate is the recommended first-line treatment and anchor drug in the management of rheumatoid arthritis.23,24 It is also effective in a wide range of other systemic inflammatory diseases and was a natural candidate for study as a treatment in GCA. 25 There has been conflicting evidence of methotrexate’s efficacy in GCA. Two randomized controlled trials (RCTs) showed no effect, whereas one reported reduced relapse rates and lower glucocorticoid doses with methotrexate. All of these studies had a number of potential limitations including the small number of patients recruited, ranging from 21 to 98 patients.26–28 An individual patient data meta-analysis of the three studies reported lower relapse rates (hazard ratio 0.65, p = 0.04), lower cumulative glucocorticoid doses (mean –842 mg at 48 weeks), and a higher rate of glucocorticoid-free remission (hazard ratio 2.8, p = 0.001) with methotrexate. 8 The evidence of efficacy from this meta-analysis has to be tempered by the realization of the relatively high numbers needed to treat (10 to prevent one cranial relapse of GCA) and the lack of evidence of a decrease in adverse events with its use. 8 In clinical practice, methotrexate is unlikely to be sufficient to result in a meaningful benefit for the majority of GCA patients.
Synthetic immunosuppressants
Other synthetic immunosuppressants, including azathioprine, leflunomide, mycophenolate mofetil, hydroxychloroquine, dapsone and cyclophosphamide, have also been used in GCA. However, the evidence supporting their use is largely limited to case series.29–35 One small non-randomized double-blind study of azathioprine in patients with either PMR or GCA showed a significant reduction in mean steroid dose over 52 weeks. 36 An RCT of hydroxychloroquine published in abstract form showed no evidence of efficacy. 37 Cyclosporin A did not demonstrate a significant steroid-sparing effect in two randomized open-label studies.38,39
Why has it been so difficult to find an effective treatment for GCA?
The explanation for the difficulties in finding an effective treatment for GCA is multifaceted. Factors such as the relative rarity of the disease and the limited extent of research interest in the area, with a small number of groups of dedicated active researchers, have played their part. However, the factors involved run deeper than this. To a large extent, until recently the majority of treatments were repurposed from other rheumatic diseases, particularly rheumatoid arthritis. While there are certain similarities between the diseases, it is perhaps not overly surprising that many of these treatments did not translate to what is a distinct disease area. An interacting and even more important factor relates to the underlying pathogenesis of GCA.
Pathogenesis of GCA
The pathogenesis of GCA remains to be fully elucidated and significant work is ongoing in this area. Despite our evolving knowledge, what has become increasingly clear is that the processes and pathways involved are complex, adding an extra level of difficulty in finding an effective treatment option. The current hypothesis of GCA pathogenesis implicates dual T-lymphocyte pathways, illustrated in Figure 1. The full discussion of GCA pathogenesis is beyond the scope of the current article and we direct interested readers to previously published reviews.6,7,40,41 This is an added complication as, if this hypothesis is correct, a truly effective treatment approach will either need to target both pathways with a single agent, or alternatively will require a combination of two agents. Fortuitously, existing biologic agents are available which have the potential to target both limbs of this pathogenic model. We will now proceed to discuss potential biologic treatment options in GCA, with particular reference to those that target the pathways implicated in the pathogenic model, namely tocilizumab (interleukin-6), abatacept (T-lymphocytes), and ustekinumab (interleukin-12/interleukin-23).

Proposed pathogenic model in GCA.
Biologic agents
Biologic agents have revolutionized the treatment of many systemic rheumatic diseases. They have provided an effective treatment option to many patients with previously intractable disease. When utilized appropriately they also reduce disability and improve capacity to work and quality of life. However, the translation of these agents and their benefits to GCA has not been a smooth one. An overview of the current biologic treatment options assessed in GCA is shown in Table 1.
Biologic agents in giant cell arteritis.
GC, glucocorticoids, GCA, giant cell arteritis; RCT, randomized controlled trial.
Tumour necrosis factor alpha inhibitors
Tumour necrosis factor alpha (TNF-α) inhibitors were the first biologic agents to be evaluated in GCA. High tissue levels of TNF-α led to speculation that anti-TNF-α agents may be effective in GCA. 42 Subsequent RCTs failed to support this theorized benefit. An RCT of infliximab in 44 newly diagnosed GCA patients who were in glucocorticoid-induced remission demonstrated no significant benefit to infliximab use and the study was halted after an interim analysis. Indeed, while not reaching statistical significance both relapse rates and infection risk were higher in the infliximab group. 10 Adalimumab was also demonstrated to have no benefit in an RCT, with remission rates of 59% compared to 50% in the placebo group at week 26. There was a numerically lower risk of adverse events in the adalimumab group in this study but this did not reach clinical or statistical significance. 9 A small RCT of the soluble TNF receptor fusion protein etanercept in 17 patients suggested that this may potentially be a more effective method of targeting TNF-α in this population than with a monoclonal antibody. A numerically greater amount of etanercept treated patients (50% versus 12%) were in glucocorticoid-free remission at 12 months and the cumulatively prednisolone dose was significantly lower in this group. 43 However, the small size of this study and the lack of any subsequent large-scale studies of etanercept mean that the results must be interpreted with caution. Overall, the weight of evidence suggests that TNF-α inhibitors are ineffective in GCA.
Tocilizumab
Tocilizumab is a monoclonal antibody against the interleukin-6 receptor and is widely utilized in rheumatoid arthritis. Interleukin-6 has emerged as an attractive therapeutic target in GCA. This is based on reports of increased levels of IL-6 mRNA expression in inflamed temporal arteries and of elevated serum IL-6 levels in patients with active GCA.42,44,45 However, this must be balanced against conflicting evidence from translational research showing that IL-6 does not stimulate pro-inflammatory cytokines in an ex vivo temporal artery biopsy model. 46 It is also concerning that lower IL-6 levels have been shown to have a clinical association with cranial ischaemic complications. 47 Coupled with our knowledge of the hypothesized dual T-cell pathway in GCA, this raises concerns that IL-6 blockade may address the TH17 cell pathway but leave unchecked the TH1 cell pathway. 48 Recent translational work, however, has shown that IL-6 may exert effects on the T-regulatory cell population in GCA, normalizing the phenotype and function of these cells in GCA. 49 This suggests that IL-6 blockade has the potential to exert effects outside of the direct targeting of the TH17 cell pathway. Work from our group has uncovered an additional potential role for IL-6 as an effector cytokine in the TH1 pathway. 41 Two RCTs of tocilizumab in GCA have reported extremely encouraging data regarding reduction in relapse rates and cumulative glucocorticoid doses. In the phase II study by Villiger and colleagues, 85% of tocilizumab patients achieved relapse-free survival at week 52 compared to 20% in the placebo group. This significant reduction in relapses was accompanied by a significant reduction in cumulative glucocorticoid dose; 43 mg/kg versus 110 mg/kg at week 52. 50 In the phase III GIACTA study, sustained remission rates at week 52 were 53% in tocilizumab-treated patients compared to 17% in glucocorticoid monotherapy, while the cumulative glucocorticoid dose was reduced by 50% in tocilizumab-treated patients. 51 The tocilizumab-treated patients in both of these studies received a shorter glucocorticoid duration than would commonly be used in clinical practice; while the 52-week glucocorticoid monotherapy arm of the GIACTA study is more comparable to standard glucocorticoid regimes there still exists a lack of clarity regarding the optimum glucocorticoid regime to use in combination with tocilizumab. It would provide further reassurance if the limited imaging performed as part of these two studies demonstrated radiographic improvement of vasculitis. The MR angiography (MRA) results from the phase II trial have been reported and demonstrated incomplete suppression of MRA evidence of vasculitis, the clinical relevance of which remains to be determined. 52 The concerns raised from translational work regarding the potential for unchecked vasculitis and ischaemic complications to occur in clinically well patients is supported by some evidence from uncontrolled clinical studies. These include widespread vasculitis in a tocilizumab-treated patient in the open-label study by Unizony and colleagues, and evidence of progression of large vessel vasculitis in two Takayasu’s arteritis patients treated with tocilizumab despite clinical and serological improvement.53,54 Further data from the two RCTs are awaited. An RCT of an IL-6 monoclonal antibody, sirukumab (ClinicalTrials.gov identifier: NCT02531633), commenced recruitment but has now been cancelled.
Abatacept
Abatacept is a T-cell modulator, an effect achieved through inhibition of co-stimulation. It does this by binding to the CD80/CD86 molecule, preventing it from providing the second signal required for T-cell activation. Abatacept is theoretically attractive in GCA as both limbs of the proposed pathogenic model are helper T-cell based, TH1 and TH17 respectively playing central roles. In an RCT in GCA, abatacept in combination with glucocorticoids resulted in an increase in relapse-free survival at 12 months, from 31% to 48%, compared to glucocorticoid monotherapy. 55 Interpreting these results is complicated by the trial design, in which all patients received abatacept for the first 12 weeks, followed by randomization to continuing or withdrawing abatacept after this point. The small improvement in outcome in GCA patients coupled with the lack of efficacy in an RCT in Takayasu’s arteritis is discouraging, and a planned phase III RCT has been withdrawn. 56 Therefore, at the present time there exists preliminary RCT evidence that abatacept is effective in GCA, but this effect appears to be modest.
Ustekinumab
Ustekinumab is a human IgG1 kappa monoclonal antibody developed using human Ig transgenic mice. 57 Ustekinumab blocks both IL-12 and IL-23 activity by binding to the common p40 subunit shared by both cytokines.58,59 Ustekinumab was originally developed with the sole goal of targeting IL-12, prior to the recognition of the functional importance of IL-23. 58 It binds to IL-12 and IL-23 equally, preventing binding to their receptor complexes on the surfaces of T-lymphocytes and NK cells. 57 Thus, although ustekinumab was initially developed to target IL-12, it has emerged as a first-in-class dual cytokine inhibitor. 58 Ustekinumab is functionally unable to bind to IL-12 or IL-23 that have already bound to receptors and so is unlikely to participate in Fc effector functions such as antibody-dependent cell-mediated cytotoxicity. 57 Ustekinumab is FDA- and EMA-approved for use in the treatment of psoriasis and psoriatic arthritis. 60 This followed evidence of efficacy from phase III trials in both diseases.61–64 It has also demonstrated clinical efficacy in the treatment of refractory Crohn’s disease. 65 The hypothesized dual role of IL-12 and IL-23 in GCA makes ustekinumab a potentially attractive treatment option in this disease. A prospective open-label study of ustekinumab was performed in 25 patients with GCA.66,67 All patients had refractory disease, defined as an inability to taper glucocorticoids below a certain fixed dose, or multiple relapses on tapering glucocorticoids. They had also failed a median of one previous steroid-sparing agent. Ustekinumab appeared to be efficacious in treating these patients, with a reduction in median prednisolone dose from 20 mg to 5 mg (p < 0.001), and a significant reduction in median C-reactive protein from 12.9 mg/L to 6 mg/L (p = 0.006). No patients had a flare of GCA while treated with ustekinumab, and one-quarter of patients were able to stop glucocorticoids entirely. CT angiography demonstrated a radiographic correlate of the clinical findings with improvement in radiographic large vessel vasculitis in all cases and full resolution in 50% of cases. There were no unexpected adverse events seen in the ustekinumab-treated patients. Three patients discontinued ustekinumab over the course of the study due to adverse events, with two of these subsequently having polymyalgic flares of GCA at 4 and 5 months respectively after stopping ustekinumab. The results from this study are certainly encouraging; however, it was an open-label study with no control group, and therefore an RCT of ustekinumab in GCA is warranted before it can be recommended for routine use.
Other agents
A number of other agents have been utilized in GCA and reported in case reports and case series. The successful use of interleukin-1β blockade with anakinra has been reported in three patients with GCA. 68 An RCT of anakinra (ClinicalTrials.gov identifier: NCT02902731) is currently planned. An RCT of another agent targeting interleukin-1β, gevokizumab, was commenced but subsequently cancelled. Efficacy of the anti-CD20 agent rituximab in refractory GCA has been reported in two case reports; while the pathogenic model of GCA mainly emphasizes the importance of T-cells, there are a number of studies that support a potential role for B-cells in GCA.69–71
As many of the cytokines implicated in GCA, including IL-6, IL-12 and IL-23, operate through JAK-STAT pathways, JAK inhibitors are an attractive potential treatment option. There is supportive basic scientific data for the efficacy of this approach with tofacitinib. 72 A small open-label pilot study of another JAK inhibitor, baricitinib, is currently recruiting participants (ClinicalTrials.gov identifier: NCT03026504). A third JAK inhibitor, upadacitinib, is also planned for an RCT (ClinicalTrials.gov identifier: NCT03725202).
Treatment approach in GCA in 2018
Integrating this new emerging evidence on the efficacy of biologic agents into a current treatment approach in GCA requires careful consideration. At present the authors would caution against the routine use of biologics in patients with new-onset GCA. While the emerging data, especially for tocilizumab, are highly promising, we still require further information particularly regarding the long-term efficacy and safety of these agents. It is reasonable to consider the institution of biologic agents in GCA patients who relapse while tapering glucocorticoids; whether the first or second relapse should be the trigger for this is unclear at present. The data supporting the use of biologic agents in GCA are strongest for tocilizumab. The use of other biologic agents, such as abatacept or ustekinumab, should probably be reserved for research settings and for those patients who have contraindications to or relapse on tocilizumab. The authors present a purely personal treatment algorithm for GCA in Figure 2. The treatment of GCA is a dynamic area at present and given the limited evidence currently available, this treatment approach is predominantly based on expert opinion and so must be used with caution and reviewed in the light of emerging data.

Suggested treatment algorithm in GCA.
Conclusion
The treatment of GCA has been based almost entirely on glucocorticoids for nearly 70 years. In that time, multiple other agents have been used with limited success. Emerging evidence suggests that biologic agents, particularly tocilizumab, may open up new treatment approaches. Further evidence is needed before they can be recommended for routine use in patients with GCA.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Conflict of interest statement
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
