You are here

Article Text

Article menu

Download PDFPDF

Vasculitis
Original research
Diagnostic accuracy of serum biomarkers to identify giant cell arteritis in patients with polymyalgia rheumatica
  1. André Ramon1,2,
  2. Hélène Greigert2,3,
  3. Karine Goueslard4,
  4. Claudie Cladière2,
  5. Marion Ciudad2,
  6. Paul Ornetti1,5,6,
  7. Sylvain Audia2,7,
  8. Jean Francis Maillefert1,6,
  9. Bernard Bonnotte2,7 and
  10. Maxime Samson2,7
  1. 1Rheumatology, University Hospital Centre Dijon Bourgogne, Dijon, France
  2. 2EFS, INSERM, UMR RIGHT, Franche-Comté University, Besançon, France
  3. 3Vascular Medicine Department, University Hospital Centre Dijon Bourgogne, Dijon, France
  4. 4Methodological Support Unit (USMR), University Hospital Centre Dijon Bourgogne, Dijon, France
  5. 5INSERM, CIC 1432, Clinical Investigation Center, Plurithematic Module, Technological Investigation Platform, Dijon-Bourgogne University Hospital, Dijon, France
  6. 6INSERM UMR 1093-CAPS, Bourgogne Franche-Comté University, Dijon, France
  7. 7Internal Medicine and Clinical Immunology Department, Referral Center for Rare Systemic Autoimmune and Autoinflammatory Diseases, University Hospital Centre Dijon Bourgogne, Dijon, France
  1. Correspondence to Dr André Ramon; andre.ramon{at}chu-dijon.fr

Abstract

Introduction Polymyalgia rheumatica (PMR) and giant cell arteritis (GCA) are frequently overlapping conditions. Unlike in GCA, vascular inflammation is absent in PMR. Therefore, serum biomarkers reflecting vascular remodelling could be used to identify GCA in cases of apparently isolated PMR.

Materials and methods 45 patients with isolated PMR and 29 patients with PMR/GCA overlap were included. Blood samples were collected before starting glucocorticoids for all patients. Serum biomarkers reflecting systemic inflammation (interleukin-6 (IL-6), CXCL9), vascular remodelling (MMP-2, MMP-3, MMP-9) and endothelial function (sCD141, sCD146, ICAM-1, VCAM-1, vWFA2) were measured by Luminex assays.

Results Patients with GCA had higher serum levels of sCD141 (p=0.002) and CXCL9 (p=0.002) than isolated PMR. By contrast, serum levels of MMP-3 (p=0.01) and IL-6 (p=0.004) were lower in GCA than isolated PMR. The area under the curve (AUC) was calculated for sCD141, CXCL9, IL-6 and MMP-3. Separately, none of them were >0.7, but combinations revealed higher diagnostic accuracy. The CXCL9/IL-6 ratio was significantly increased in patients with GCA (p=0.0001; cut-off >32.8, AUC 0.76), while the MMP-3/sCD141 ratio was significantly lower in patients with GCA (p<0.0001; cut-off <5.3, AUC 0.79). In patients with subclinical GCA, which is the most difficult to diagnose, sCD141 and MMP-3/sCD141 ratio demonstrated high diagnostic accuracy with AUC of 0.81 and 0.77, respectively.

Conclusion Combined serum biomarkers such as CXCL9/IL-6 and MMP-3/sCD141 could help identify GCA in patients with isolated PMR. It could allow to select patients with PMR in whom complementary examinations are needed.

  • Giant Cell Arteritis
  • Polymyalgia Rheumatica
  • Vasculitis

Data availability statement

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/rmdopen-2024-004488

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Polymyalgia rheumatica (PMR) and giant cell arteritis (GCA) are two frequent overlapping conditions.

    • Recent studies have reported a prevalence of 10–30% of subclinical GCA in new-onset PMR.

    • Routine inflammatory markers (C reactive protein, erythrocyte sedimentation rate) have poor diagnostic accuracy to identify patients with GCA.

    WHAT THIS STUDY ADDS

    • Serum MMP-3/sCD141 and CXCL9/IL-6 ratios could be effective in identifying GCA in patients with PMR.

    • In subclinical GCA, sCD141 and MMP-3/sCD141 ratio kept a high diagnostic accuracy.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • These serum biomarkers could help clinicians to select patients with PMR in whom complementary examinations are needed.

    • Prospective studies are needed to confirm these results with a greater number of patients with subclinical GCA in an independent cohort.

    Introduction

    Polymyalgia rheumatica (PMR) is an inflammatory rheumatic disease characterised by inflammatory pain of the pelvic and scapular girdles (bursitis and/or tenosynovitis) that occurs in individuals older than 50 years.1 In 16–21% of cases, PMR is associated with giant cell arteritis (GCA), a large-vessel (LV) vasculitis that mainly affects the aorta and its branches, and which is the most common vasculitis in individuals over 50 years.2 In cases of PMR, associated GCA is easily suspected in patients with cranial signs (temporal headache, scalp tenderness, jaw claudication, visual impairment). However, the diagnosis may be more challenging, particularly in the case of extracranial involvement (LV GCA), due to the lack of specific clinical signs in this type of patients.3 Furthermore, various studies have demonstrated that 10–30% of patients with isolated PMR symptoms have subclinical GCA.4 Accordingly, it seems important to identify serum biomarkers that can detect GCA in individuals with PMR, in order to select patients warranting complementary investigations (temporal artery biopsy (TAB), LV imaging examinations).

    In GCA, inflammation occurs within the walls of large-sized arteries, leading to vascular remodelling and signs of ischaemia.5 The in situ production of interferon gamma (IFN-γ) by Th1 cells leads to the activation of vascular smooth muscle cells (VSMC), which further produce CCL2, CXCL9, CXCL10 and CXCL11.6 This results in the recruitment of additional Th1 cells, CD8+ T cells and CCR2+ monocytes, which differentiate into macrophages producing high amounts of interleukin-6 (IL-6) and matrix metalloproteinase (MMP).7 8 IL-6 plays a major role in systemic inflammation,9 whereas MMPs, mainly MMP-2 and MMP-9, produced by both macrophages and VSMCs, are involved in vascular remodelling by destroying cellular matrix proteins, and induce media destruction and internal elastic lamina (IEL) fragmentation, which is specific to GCA when compared with PMR.10 Endothelial cells are also involved in GCA pathogenesis. T cells are thought to be recruited via vasa vasorum of the adventitia, in which endothelial cells express adhesion molecules such as intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1).11 Furthermore, previous studies have demonstrated higher concentrations of soluble thrombomodulin (sCD141) in patients with Takayasu arteritis and GCA.12 Thrombomodulin is a transmembrane protein expressed on the surface of endothelial cells that play a key role in preventing inflammatory infiltration of the endothelium.13 14 More recently, in order to discriminate GCA from PMR, studies of serum biomarkers have also demonstrated that the serum levels of angiopoietin-2 and MMP-315 16 were predictive of GCA in a population of patients with PMR.

    Therefore, we hypothesised that biomarkers reflecting vascular inflammation and/or remodelling could help identify GCA in patients with apparently isolated PMR. We explored this hypothesis by comparing patients with isolated PMR and patients with PMR/GCA overlap.

    Materials and methods

    Patients

    Patients were prospectively enrolled in the rheumatology and internal medicine departments of the Dijon-Bourgogne University Hospital with a systematic protocol of work-up allowing assessment of cranial and extracranial vessels. These patients were referred by a general practitioner or private rheumatologist based on suspicion of GCA, that is, patients with PMR features associated with cranial signs and/or constitutional signs and/or significant inflammatory syndrome. All included patients fulfilled the PMR classification criteria from the American College of Rheumatology (ACR) and the European Alliance of Associations for Rheumatology (EULAR) 2012.17 The ‘isolated PMR group’ was composed of patients with new-onset isolated PMR, without GCA. A 6-month follow-up period ensured that there were no alternative diagnoses. Furthermore, we systematically excluded an association with GCA at baseline by performing an evaluation of temporal arteries by colour Doppler ultrasound (CDUS) and/or TAB and also an assessment of LVs by [18F]-fluorodeoxyglucose-positron emission tomography/CT (18FDG-PET/CT) or angio-CT scan (online supplemental table 1). Patients with PMR, for whom a proof of vasculitis was obtained by TAB or vascular imaging at baseline and who fulfilled the ACR/EULAR 2022 GCA classification criteria,18 were grouped in the GCA group. Positive TAB was defined by a non-necrotising granulomatous panarteritis, with an inflammatory cellular infiltrate composed of mononuclear cells (T lymphocytes and macrophages) and/or giant cells with fragmentation of the IEL, and destruction of the media.2 The 18FDG-PET/CT assessment was visually scored from 0 to 3, where 0=no FDG uptake; 1=low-grade uptake (less than liver uptake); 2=intermediate-grade uptake (similar to liver uptake); and 3=high-grade uptake (higher than liver uptake). GCA was ruled out in patients with a score of ≤1 on vascular segments.19 All 18FDG-PET/CT was performed within a period of <3 days if glucocorticoids were introduced. For patients with angio-CT scan assessment, GCA diagnosis was retained in case of circumferential arterial parietal thickening (>2 mm for aortitis).20

    Serum biomarkers

    Blood samples were collected at diagnosis of PMR or GCA for all patients who were free of glucocorticoids or immunosuppressive treatment at the time of biological sampling. Serum was isolated after centrifugation at 3600 rpm for 10 min within 2 hours after collection. Serum was then stored at −80°C for subsequent assays. Serum levels of soluble (s)CD141 (standard curve: 86.4–21 000 pg/mL; sensitivity (se): 7.06 pg/mL), sCD146 (185–45 000 pg/mL; 3.50 pg/mL), von Willebrand factor (vWFA2) (41.2–10 000 pg/mL; 11.9 pg/mL), MMP-2 (276–67 000 pg/mL; 108 pg/mL), MMP-3 (82.3–20 000 pg/mL; 5.3 pg/mL), MMP-9 (123–30 000 pg/mL; 13.6 pg/mL), ICAM-1 (7000–1 700 000 pg/mL; 87.9 pg/mL), VCAM-1 (8850–2 150 000 pg/mL; 238 pg/mL), IL-6 (3.25–790 pg/mL; 1.7 pg/mL), CXCL9 (617–150 000 pg/mL; 23.8 pg/mL) and sCD163 (5140–1 250 000 pg/mL; 530 pg/mL) were measured using a Human Premixed Magnetic Luminex screening assay kit (R&D Systems) according to manufacturer’s instructions, and read and analysed on Bio-Plex 200 System assay (Bio-Rad, France).

    Ethics

    All included patients signed written informed consent in accordance with the Declaration of Helsinki.

    Statistics

    Data are expressed as numbers (%) for categorical variables and medians (IQR) for continuous variables. Non-parametric tests (Mann-Whitney) were performed to compare data, as appropriate. The diagnostic performance of serum biomarkers was assessed by analysing the receiver operating characteristic with the area under the curve (AUC). Optimal cut-off points were identified by assessing the maximum of the sum of sensitivity and specificity, according to the Youden Index. Spearman’s rank correlation coefficient was performed to assess the relationship between serum biomarkers. The significance threshold was set at p<0.05 (two tailed). Analyses were performed with GraphPad PRISM software.

    Results

    Patient characteristics

    In total, 45 patients with PMR and 29 patients with GCA were included. Median age, sex and body mass index at baseline were similar in the two groups. Among the patients with GCA, 21 (72.4%) had cranial symptoms, mainly temporal headache (55.1%), scalp tenderness (37.9%) and jaw claudication (34.4%). Eight patients with GCA had no clinical sign of GCA at baseline (ie, subclinical GCA). 12 patients with GCA (41.4%) had aortitis according to 18FDG-PET/CT or angio-CT scan, and 16 (55.1%) had a temporal halo sign on ultrasound.

    Among the 45 patients with PMR, TAB was performed in 38 patients and was always negative. 23 patients underwent 18FDG-PET/CT and 26 underwent angio-CT scan (four patients had both 18FDG-PET/CT and angio-CT scan). Patient characteristics are summarised in online supplemental table 2.

    Serum biomarker accuracy in patients with GCA

    No significant differences were found between patients with PMR and GCA for the levels of haemoglobin, leucocytes, neutrophils, lymphocytes, platelets, C reactive protein (CRP) and erythrocyte sedimentation rate (ESR) (online supplemental figure 1, online supplemental table 3). Patients with GCA had significantly higher serum levels of sCD141 (6.9 ng/mL (5.1–8.1) vs 5.5 ng/mL (4.4–6.3), p=0.002) and CXCL9 (565.9 pg/mL (411.3–958.3) vs 402 pg/mL (402.0–572.9), p=0.002) than patients with isolated PMR. Conversely, serum levels of MMP-3 (39.9 ng/mL (27.3–55.3) vs 28.6 ng/mL (10.1–45.3), p=0.01) and IL-6 (16.1 pg/mL (12.4–34.7) vs 10.6 pg/mL (7.4–17.5), p=0.004) were found to be significantly higher in patients with isolated PMR compared with patients with GCA (figure 1A, online supplemental table 3). No significant differences were found between patients with isolated PMR and GCA for sCD146 (p=0.21), ICAM-1 (p=0.05), VCAM-1 (p=0.14), vWFA2 (p=0.08), MMP-2 (p=0.20), MMP-9 (p=0.34) and sCD163 (p=0.47) (online supplemental figure 1, online supplemental table 3).

    Figure 1
    Figure 1

    Concentration level of serum biomarkers and receiver operating characteristic (ROC) curve in patients with isolated PMR, GCA and subclinical GCA. (A) Concentration level and ratio of sCD141, MMP-3, IL-6, CXCL9, MMP-3/sCD141 and CXCL9/IL-6 in patients with isolated PMR, GCA and subclinical GCA. Bars represent the median. (B) ROC curve of sCD141, MMP-3/sCD141 and CXCL9/IL-6 in patients with GCA and subclinical GCA. *P<0.05; **p<0.005; ***p<0.0005; ****p<0.0001. AUC, area under the curve; GCA, giant cell arteritis; IL-6, interleukin-6; MMP, matrix metalloproteinase; NS, non significant; PMR, polymyalgia rheumatica; SUB GCA, subclinical GCA.

    For biomarkers with significantly different serum levels between the two groups (sCD141, MMP-3, IL-6, CXCL9), none demonstrated good diagnostic performance separately (all AUC ≤0.7) (online supplemental table 3). However, some combinations (CXCL9/IL-6 ratio and MMP-3/sCD141 ratio) demonstrated better diagnostic performance for identifying GCA: AUC 0.76 (cut-off >32.8; se=0.79; specificity (spe)=0.73) for the CXCL9/IL-6 ratio and AUC 0.79 (cut-off <5.3; se=0.79; spe=0.71) for the MMP-3/sCD141 ratio (table 1, figure 1B).

    View this table:
    Table 1

    Diagnostic accuracy of isolated or combined sCD141, MMP-3, IL-6 and CXCL9 in the whole study population and subclinical GCA

    Serum biomarker accuracy in patients with subclinical GCA

    In the GCA group, eight patients had subclinical GCA at baseline, defined as the absence of equivocal clinical signs of GCA. For these patients, the diagnostic of GCA was obtained by 18FDG-PET or angio-CT scan showing LV vasculitis (aortitis) in five patients and TAB or CDUS which was positive in five patients (two patients had both positive TAB and positive 18FDG-PET). Between patients with subclinical GCA and isolated PMR, serum levels of sCD146 (p=0.17), ICAM-1 (p=0.12), VCAM-1 (p=0.29), vWFA2 (p=0.68), MMP-2 (p=0.89), MMP-3 (p=0.53), MMP-9 (p=0.93) and sCD163 (p=0.59) were similar as usual inflammatory markers (CRP, ESR, leucocytes) (online supplemental table 4, online supplemental figure 1). Patients with subclinical GCA had higher sCD141 concentrations than patients with isolated PMR (7.9 ng/mL (6.0–11.2) vs 5.5 ng/mL (4.4–6.3), p=0.004). In this population, the AUC of sCD141 and MMP-3/sCD141 ratio was 0.81 (se=0.75, spe=0.87) and 0.77 (se=0.75, spe=0.73), respectively. Serum levels of IL-6 and CXCL9 were not significantly different between PMR and subclinical GCA, and the CXCL9/IL-6 ratio had poor diagnostic performance with an AUC of 0.59 (table 1, figure 1A,B).

    Correlation between biomarkers

    Correlations between the different biomarkers were assessed in patients with PMR and GCA (figure 2). In patients with GCA, sCD141 was positively correlated with sCD146 (r=0.57; p=0.001), MMP-2 (r=0.46; p=0.01), MMP-3 (r=0.57; p=0.001) and VCAM-1 (r=0.47; p=0.01). IL-6 was correlated with CRP (r=0.49; p=0.006) and sCD163 was positively correlated with CXCL9 (r=0.48; p=0.009). In isolated PMR, sCD141 was positively correlated with sCD146 (r=0.37; p=0.01), sCD146 was negatively correlated with CRP (r=−0.37; p=0.01) and there was a lower correlation between CRP and IL-6 (r=0.32; p=0.01) than in patients with GCA.

    Figure 2
    Figure 2

    Correlations of serum biomarkers in patients with GCA and isolated PMR. Correlation coefficients between serum biomarkers in patients with GCA and isolated PMR. Colours indicate the strength of the association. Statistical significance is indicated as *p<0.05, **p<0.01, ***p<0.005. CRP, C reactive protein; GCA, giant cell arteritis; ICAM-1, intercellular adhesion molecule-1; IL-6, interleukin-6; MMP, matrix metalloproteinase; PMR, polymyalgia rheumatica; VCAM-1, vascular cell adhesion molecule-1; vWFA2, von Willebrand factor.

    Discussion

    This pilot study evaluated the ability of serum biomarkers to identify GCA in a population of patients with PMR. Because vascular inflammation is specific to GCA and absent in isolated PMR, we focused on biomarkers reflecting and/or related to the vascular remodelling process as it is thought to be specific of GCA. Some of the serum biomarkers evaluated here seem promising: sCD141, MMP-3, IL-6 and CXCL9 were indeed significantly different between patients with isolated PMR and GCA.

    Then, sCD141 (or soluble thrombomodulin) could be a biomarker of interest to identify GCA either combined with MMP-3 or alone for subclinical GCA by reflecting endothelial aggression with a release of soluble fragment into circulation, as demonstrated in previous studies.12 Another biomarker of interest was MMP-3. There are several hypotheses for the high MMP-3 concentrations during isolated PMR. First, synovial tissue is known to release high levels of MMP-3.21 However, in the present study, elevated MMP-3 levels during PMR do not seem to be due to synovial tissue production because patients in both groups had PMR. Second, it has been hypothesised that IFN-γ inhibits MMP-3 production.22 Last, the decrease in serum levels of MMP-3 could be explained by its consumption by activated MMP-9, which is involved in vascular remodelling process during GCA.23 In the present study, serum IL-6 was higher in patients with isolated PMR. We believe this can be explained in two ways. First, there is good evidence that patients with GCA with a cranial phenotype, especially in cases of ischaemic complication, have a lower serum IL-6 concentration.24 In addition, patients with isolated PMR were included in tertiary centres. These were, therefore, severe cases of PMR with high levels of inflammation, which is not always the case in outpatient rheumatology practice. In this study, patients with GCA were characterised by higher serum levels of CXCL9, which is released by VSMC following exposure to IFN-γ.5 Therefore, we think that this result could reflect the activation of VSMCs, and the subsequent vascular remodelling process that is observed in GCA but not in isolated PMR. The CXCL9/IL-6 ratio thus could reflect the ratio between vascular remodelling activation and systemic inflammation. Seeing as there is no vascular remodelling in isolated PMR, there is an imbalance between IL-6 and CXCL9 concentrations, whereas the higher ratio in GCA could reflect dominant vascular remodelling, while the lower level of IL-6 could be explained by the mainly cranial phenotype in our GCA population.

    Interestingly, in subclinical GCA, CXCL9 and CXCL9/IL-6 ratio had poor diagnostic accuracy in contrast to sCD141. One hypothesis could be that sCD141 serum levels could reflect an earlier stage of vascular remodelling before the involvement of VSMCs that lead to parietal thickening and the cranial manifestations of GCA reflecting by CXCL9 even functional studies are needed to confirm this hypothesis.

    This study has some limitations. The first was the small number of patients recruited, particularly the population of interest, that is, patients with subclinical GCA, which is the most interesting population in routine care in whom to perform these assays. Second, all patients included in this study were referred to our centre because they had PMR features, and the clinician suspected possible GCA due to suggestive clinical signs (temporal headache, scalp tenderness, jaw claudication, etc) and/or constitutional signs and/or an inflammatory syndrome deemed too high for a simple PMR. Therefore, these results do not apply to all patients with PMR, as patients with PMR in this study are likely to be more inflammatory than those seen in primary care. This could explain the higher IL-6 level observed in the PMR patient group. However, it is in this patient population that biomarkers are needed to rule out GCA. Despite these limitations, this study has some strengths. First, all patients were free of glucocorticoids at inclusion. Furthermore, subclinical GCA was systematically ruled out in patients with PMR with a TAB, imaging examinations or both. Finally, a 6-month retrospective follow-up period was used to ensure the absence of differential diagnoses in the PMR population.

    Conclusion

    This exploratory study demonstrates a potential interest of CXCL9/IL-6 and MMP-3/sCD141 ratios for identifying GCA in patients with PMR. Identifying subclinical GCA in patients with PMR can be particularly difficult. In this situation, serum levels of sCD141 and the MMP-3/sCD141 ratio were found to have a high level of sensitivity and specificity for GCA in this study, suggesting that they could be used to select the patients with PMR requiring complementary examinations. These exploratory results need to be confirmed prospectively in a larger number of patients, especially those with a subclinical GCA phenotype, and further validated in an independent cohort.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants and was approved by the local ethics committee. Collection of serum was declared to the French Ministry of Health (DC-2013-1933). Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    We thank Dr Anabelle Legrand, Nicolas Pernet and Serge Monier from the flow cytometry core facility (INSERM UMR1231, Université de Bourgogne Franche-Comté) for their technical support. We thank Marion Laboz, Marine Vantard, Christine Piroth, Leila Benguella (Department of Rheumatology, Dijon University Hospital), Sabine Berthier, Vanessa Leguy-Seguin, Jérôme Razanamahery, Caroline Raymond and Sethi Ouandji (Department of Internal Medicine and Clinical Immunology, Dijon University Hospital) for the recruitment of patients. We thank Suzanne Rankin for proofreading the article.

    References

    1. ↵
      1. Buttgereit F,
      2. Dejaco C,
      3. Matteson EL, et al
      . Polymyalgia rheumatica and giant cell arteritis: a systematic review. JAMA 2016;315:2442–58. doi:10.1001/jama.2016.5444
      OpenUrlCrossRefPubMed
    2. ↵
      1. Samson M,
      2. Corbera-Bellalta M,
      3. Audia S, et al
      . Recent advances in our understanding of giant cell arteritis pathogenesis. Autoimmun Rev 2017;16:833–44. doi:10.1016/j.autrev.2017.05.014
      OpenUrlPubMed
    3. ↵
      1. Dejaco C,
      2. Duftner C,
      3. Buttgereit F, et al
      . The spectrum of giant cell arteritis and polymyalgia rheumatica: revisiting the concept of the disease. Rheumatol (Oxford) 2017;56:506–15. doi:10.1093/rheumatology/kew273
      OpenUrlCrossRefPubMed
    4. ↵
      1. Hemmig AK,
      2. Gozzoli D,
      3. Werlen L, et al
      . Subclinical giant cell arteritis in new onset polymyalgia rheumatica a systematic review and meta-analysis of individual patient data. Semin Arthritis Rheum 2022;55:152017. doi:10.1016/j.semarthrit.2022.152017
      OpenUrl
    5. ↵
      1. Greigert H,
      2. Genet C,
      3. Ramon A, et al
      . New insights into the pathogenesis of giant cell arteritis: mechanisms involved in maintaining vascular inflammation. J Clin Med 2022;11:2905. doi:10.3390/jcm11102905
    6. ↵
      1. Weyand CM,
      2. Goronzy JJ
      . Immune mechanisms in medium and large-vessel vasculitis. Nat Rev Rheumatol 2013;9:731–40. doi:10.1038/nrrheum.2013.161
      OpenUrlCrossRefPubMed
    7. ↵
      1. Corbera-Bellalta M,
      2. Alba-Rovira R,
      3. Muralidharan S, et al
      . Blocking GM-CSF receptor α with mavrilimumab reduces infiltrating cells, pro-inflammatory markers and neoangiogenesis in ex vivo cultured arteries from patients with giant cell arteritis. Ann Rheum Dis 2022;81:524–36. doi:10.1136/annrheumdis-2021-220873
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Samson M,
      2. Ly KH,
      3. Tournier B, et al
      . Involvement and prognosis value of CD8(+) T cells in giant cell arteritis. J Autoimmun 2016;72:73–83. doi:10.1016/j.jaut.2016.05.008
      OpenUrl
    9. ↵
      1. Roche NE,
      2. Fulbright JW,
      3. Wagner AD, et al
      . Correlation of interleukin-6 production and disease activity in polymyalgia rheumatica and giant cell arteritis. Arthritis Rheum 1993;36:1286–94. doi:10.1002/art.1780360913
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Watanabe R,
      2. Maeda T,
      3. Zhang H, et al
      . MMP (Matrix Metalloprotease)-9-Producing Monocytes Enable T Cells to Invade the Vessel Wall and Cause Vasculitis. Circ Res 2018;123:700–15. doi:10.1161/CIRCRESAHA.118.313206
      OpenUrlPubMed
    11. ↵
      1. Ly K-H,
      2. Régent A,
      3. Tamby MC, et al
      . Pathogenesis of giant cell arteritis: more than just an inflammatory condition? Autoimmun Rev 2010;9:635–45. doi:10.1016/j.autrev.2010.05.002
      OpenUrlCrossRefPubMed
    12. ↵
      1. Boehme MW,
      2. Schmitt WH,
      3. Youinou P, et al
      . Clinical relevance of elevated serum thrombomodulin and soluble E-selectin in patients with wegener’s granulomatosis and other systemic vasculitides. Am J Med 1996;101:387–94. doi:10.1016/S0002-9343(96)00230-6
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Loghmani H,
      2. Conway EM
      . Exploring traditional and nontraditional roles for thrombomodulin. Blood 2018;132:148–58. doi:10.1182/blood-2017-12-768994
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Conway EM,
      2. Van de Wouwer M,
      3. Pollefeyt S, et al
      . The lectin-like domain of thrombomodulin confers protection from neutrophil-mediated tissue damage by suppressing adhesion molecule expression via nuclear factor kappaB and mitogen-activated protein kinase pathways. J Exp Med 2002;196:565–77. doi:10.1084/jem.20020077
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. van Sleen Y,
      2. Therkildsen P,
      3. Nielsen BD, et al
      . Angiopoietin-2/-1 ratios and MMP-3 levels as an early warning sign for the presence of giant cell arteritis in patients with polymyalgia rheumatica. Arthritis Res Ther 2022;24:65. doi:10.1186/s13075-022-02754-5
      OpenUrl
    16. ↵
      1. Fukui S,
      2. Nunokawa T,
      3. Kobayashi S, et al
      . MMP-3 can distinguish isolated PMR from PMR with GCA: a retrospective study regarding PMR and GCA in Japan. Mod Rheumatol 2016;26:259–64. doi:10.3109/14397595.2015.1071304
      OpenUrl
    17. ↵
      1. Dasgupta B,
      2. Cimmino MA,
      3. Maradit-Kremers H, et al
      . 2012 provisional classification criteria for polymyalgia rheumatica: a European league against rheumatism/American college of rheumatology collaborative initiative. Ann Rheum Dis 2012;71:484–92. doi:10.1136/annrheumdis-2011-200329
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Ponte C,
      2. Grayson PC,
      3. Robson JC, et al
      . 2022 American college of rheumatology/EULAR classification criteria for giant cell arteritis. Ann Rheum Dis 2022;81:1647–53. doi:10.1136/ard-2022-223480
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Thibault T,
      2. Durand-Bailloud B,
      3. Soudry-Faure A, et al
      . PET/CT of cranial arteries for a sensitive diagnosis of giant cell arteritis. Rheumatology (Oxford) 2023;62:1568–75. doi:10.1093/rheumatology/keac430
      OpenUrl
    20. ↵
      1. Berthod PE,
      2. Aho-Glélé S,
      3. Ornetti P, et al
      . CT analysis of the aorta in giant-cell arteritis: a case-control study. Eur Radiol 2018;28:3676–84. doi:10.1007/s00330-018-5311-8
      OpenUrl
    21. ↵
      1. Wu X,
      2. Liu Y,
      3. Jin S, et al
      . Single-cell sequencing of immune cells from anticitrullinated peptide antibody positive and negative rheumatoid arthritis. Nat Commun 2021;12:4977. doi:10.1038/s41467-021-25246-7
    22. ↵
      1. Hu X,
      2. Ho HH,
      3. Lou O, et al
      . Homeostatic role of interferons conferred by inhibition of IL-1-mediated inflammation and tissue destruction. J Immunol 2005;175:131–8. doi:10.4049/jimmunol.175.1.131
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Johnson JL,
      2. Dwivedi A,
      3. Somerville M, et al
      . Matrix metalloproteinase (MMP)-3 activates MMP-9 mediated vascular smooth muscle cell migration and neointima formation in mice. Arterioscler Thromb Vasc Biol 2011;31:e35–44. doi:10.1161/ATVBAHA.111.225623
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Hernández-Rodríguez J,
      2. Segarra M,
      3. Vilardell C, et al
      . Elevated production of interleukin-6 is associated with a lower incidence of disease-related ischemic events in patients with giant-cell arteritis: angiogenic activity of interleukin-6 as a potential protective mechanism. Circulation 2003;107:2428–34. doi:10.1161/01.CIR.0000066907.83923.32
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors AR and MS were the principal investigators. AR, HG, SA, PO, JFM, BB and MS included the patients. AR, CC and MC performed the laboratory work. AR and KG did the statistical analyses. AR, HG, KG, BB and MS contributed to data interpretation. AR, HG, BB and MS drafted the manuscript. All authors contributed to the article and approved the submitted version. AR is the guarantor.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests MS: AbbVie consulting, Argenx consulting, Boehringer Ingelheim consulting, GSK consulting, Novartis consulting and research grant, Roche–Chugai consulting, CSL Vifor consulting, Fresenius consulting. BB: Roche–Chugai personal fees for consulting, Boehringer Ingelheim consulting. AR: Novartis research grant, Novartis consulting, AbbVie consulting, Boehringer Ingelheim consulting, Chugai consulting.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.