Figures
Abstract
Introduction
Previous studies suggest a role for eotaxin-3, TARC/CCL17 and IgG4 in newly- diagnosed patients with eosinophilic granulomatosis with polyangiitis (EGPA, Churg-Strauss) with highly active disease. The role of these biomarkers in relapsing disease is unclear.
Methods
Serum levels of TARC/CCL17, eotaxin-3, IgG4, and IgG4/IgG ratio were determined in serum samples from a longitudinal cohort of patients with EGPA (105 visits of 25 patients). Epidemiological, clinical and laboratory data were available for all visits.
Results
At the first visit, 80% of patients were using glucocorticoids and 68% additional immunosuppressive drugs. Disease flares were seen at 18 visits. The median BVAS and BVAS/WG scores at time of relapse were 4 and 2, respectively. None of the biomarkers tested were useful to discriminate between active disease and remission. Patients treated with prednisone had lower eotaxin-3 and eosinophil levels compared to patients not taking glucocorticoids irrespective of disease activity. Use of immunosuppressive agents was not associated with biomarker levels.
Citation: Dejaco C, Oppl B, Monach P, Cuthbertson D, Carette S, Hoffman G, et al. (2015) Serum Biomarkers in Patients with Relapsing Eosinophilic Granulomatosis with Polyangiitis (Churg-Strauss). PLoS ONE 10(3): e0121737. https://doi.org/10.1371/journal.pone.0121737
Academic Editor: Masataka Kuwana, Nippon Medical School Graduate School of Medicine, JAPAN
Received: August 16, 2014; Accepted: February 15, 2015; Published: March 26, 2015
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication
Data Availability: All relevant data are within the paper.
Funding: The Vasculitis Clinical Research Consortium has received support from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (U54AR057319 and U01 AR51874 04), the National Center for Research Resources (U54 RR019497); the Office of Rare Diseases Research; and the National Center for Advancing Translational Science. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Eosinophilic granulomatosis with polyangiitis (EGPA, Churg-Strauss) is a rare systemic vasculitis almost exclusively arising in asthma patients [1]. Although EGPA can occur in any organ system, the most frequent disease manifestations are confined to the respiratory tract and peripheral nervous system. Skin, gastro-intestinal tract, and cardiac involvement are also common, with the latter being the major determinant of mortality in the long- term [2]. Marked eosinophilia is an extremely common feature of EGPA. Anti-neutrophil cytoplasmic antibodies (ANCA) are present in one third of patients at diagnosis and are associated with the presence of glomerulonephritis [3]. On a microscopic level, EGPA typically causes an eosinophil- rich, granulomatous, necrotizing vasculitis [4]. Mild disease is usually treated with glucocorticoids alone, whereas severe disease requires additional therapy with immunosuppressive drugs such as cyclophosphamide [5,6].
Relapses are common in EGPA and clinical experience suggests that severe flares are usually accompanied by blood eosinophilia [7]. However, intermittent elevation of eosinophil count without change in clinical activity of EGPA itself is also frequent. Moreover, discriminating disease activity from worsening of underlying asthma and sinusitis is challenging. Thus, biomarkers supporting the diagnosis of disease flares in EGPA would be extremely helpful. In prior studies of untreated patients with EGPA with active disease, we found several potential biomarkers of disease activity including serum levels of eotaxin-3, TARC/CCL17, and IgG4 [8–10]. In the current study, we investigated a panel of biomarkers in a prospective, longitudinal EGPA cohort to determine the marker’s value in identifying relapsing disease.
Methods
Study patients
Twenty-five patients enrolled between 2006 and 2012 in the Vasculitis Clinical Research Consortium (VCRC) Longitudinal Study of EGPA were selected for this analysis. The VCRC is a multicenter research infrastructure supported by US National Institutes of Health (NIH). Patients fulfilled the 1990 American College of Rheumatology (ACR) classification criteria for Churg-Strauss syndrome [11]. All subjects were enrolled using protocols and written informed consent documents approved by local institutional review boards/ethics committees at all institutions at which participants were recruited: Boston University (Boston University Medical Campus IRB), University of South Florida (USF IRB), McMaster University (McMaster Research Ethics Board), Mount Sinai Hospital Toronto (The Mount Sinai Research Ethics Board), Cleveland Clinic (IRB Cleveland Clinic Foundation), University of Utah (University of Utah IRB), University of Pittsburgh (University of Pitts burgh IRB), Johns Hopkins University (The Johns Hopkins Medicine IRB), Mayo Clinic (Mayo Clinic Institutional Review Board) and University of Pennsylvania (University of Pennsylvania IRB).
Clinicians with expertise in vasculitis performed comprehensive clinical assessments at each visit. Features of disease directly attributable to EGPA were recorded using standardized data collection forms. Summary measures of disease activity at each visit included calculation of the Birmingham Vasculitis Activity Score (BVAS) and the BVAS for Wegener’s Granulomatosis (BVAS/WG) [12,13]. BVAS and BVAS/WG scores indicate new or worse disease activity within the prior four weeks. In addition, physician-investigators categorically rated patients’ disease activity within the past four weeks as either remission, mild, moderate or high disease activity. For statistical analyses, we pooled data from patients’ visits with mild, moderate and high disease activity into one category named “active”. All patients’ visits categorized as active revealed by BVAS/WG and BVAS. Physician global assessment of disease activity within the prior 28 days was evaluated on a 0–10 Likert scale (BVAS/WG PGA). Worsening asthma alone in the absence of involvement of other organ systems was not sufficient to classify a patient as active EGPA.
All 18 patients with EGPA in the VCRC cohort (as of 2012) who had a clinical relapse and who had additional samples taken during remission were investigated. Serum samples were retrieved at the relapse visits, and further sampling (1–7 per patient) was performed at visits conducted in phases of remission 1–8 months before and/or 3–24 months after a flare. Seven patients with EGPA without clinical relapse were studied in a similar way: 2–5 serum samples across 3–12 months.
Biomarker studies
Results of the following tests were obtained for all patients at all study visits: Erythrocyte sedimentation rate (ESR, mm/hour), C-reactive protein (CRP, mg/L), and eosinophil count were measured at the VCRC sites using standard methods. Serum levels of eotaxin-3 and TARC/CCL17 were determined by ELISA (both R&D Systems, Minneapolis, MN, USA) as previously described [8,10]. Serum levels of total IgG and IgG4 were analyzed by nephelometry.
Statistical analysis
Descriptive statistics were used to summarize data. The Kolmogorov-Smirnov test was used to check for normality. For data with a normal distribution, mean ± standard deviation is presented whereas for non-normally distributed data the median and range is reported. We used the Mann-Whitney U test for comparisons of (not normally distributed) data. Proportions were analyzed by chi-square or Fisher’s exact test, as appropriate. Statistical analysis was performed using SPSS version 21 (SPSS Inc., Chicago, IL).
Multivariate inclusive logistic regression (maximum likelihood method) models were conducted to investigate the association between biomarkers, disease activity and glucocorticoid use. There was insufficient power to perform a repeated measurement model due to the low number of patients and visits. Therefore, we analyzed each visit as if it was an independent case and adjusted for possible clustering of findings within patients including the factor patient into the regression model. We applied either disease activity (active or remission as determined by treating physicians), the use of glucocorticoids (yes or no) or treatment with immunosuppressive agents (yes or no) as variables of primary interest. Glucocorticoids, immunosuppressive agents and disease activity were included as covariates in these models. Goodness of fit was tested with Hosmer-Lemeshow statistics and for sensitivity analysis we excluded cases producing low/high DFBETAs and/or large Cook values.
Results
Patient Characteristics
We analyzed 105 visits of 25 patients with EGPA (median 4 visits per patient, range 2–8) (Table 1). Mean age of the patients was 49.7 ± 13 years with similar numbers of males and females. At their first visits, 20 patients (80%) were receiving prednisone and 17 (68%) had additional immunosuppressive therapies (9 methotrexate, 7 azathioprine and 1 mycophenolate mofetil). Remission was observed in 87 (83%) visits, while disease flares were seen at 18 (17%) visits. Five of the 18 flares observed were characterized as moderate or severe, while the other 13 were regarded as mild. No patient had more than one flare. Median BVAS at flare visits was 4 (range 0–18) and median BVAS/WG was 2 (range 1–4). EGPA symptoms most commonly observed in relapsing patients were: constitutional symptoms, musculoskeletal pain, sinusitis and peripheral neuropathy. Cardiovascular events, renal and gastrointestinal disease manifestations were rare or absent during relapsing disease in this cohort.
Serum levels of eotaxin-3, TARC/CCL17 and IgG4 in patients with EGPA
Serum levels of eotaxin-3, TARC/CCL17, IgG4, and IgG4/IgG ratio did not discriminate between periods of remission and active disease within individual patients (Table 2), even after adjustment for prednisone use. Similarly, ESR, CRP, and eosinophil count also failed to differentiate between active and inactive disease.
Next, we asked whether biomarker levels differed between groups defined by use of prednisone or immunosuppressive agents. Results are detailed in Table 3 and Table 4. As expected, eosinophil counts were higher in non-users of prednisone than in users, but ESR, CRP, serum TARC/CCL17 and IgG4/IgG ratio was not significantly different. Eotaxin-3 levels were also increased in patients not on prednisone. IgG4 levels were lower in patients not taking prednisone according to unadjusted analysis; however, the difference was not significant after correction for disease activity and use of immunosuppressive agents. Similarly, we observed lower eotaxin-3 levels and a higher ESR in patients on immunosuppressive agents in unadjusted analysis, whereas in the multivariate model, the difference was not significant.
Discussion
Little is known about serologic markers of activity in relapsing patients with EGPA. Defining active disease in EGPA is especially challenging because almost all patients have underlying inflammatory disease that does not necessarily indicate activity of EGPA per se, such as asthma and symptoms of rhinosinusitis. Symptoms of asthma may recur together with signs of vasculitis in patients with active EGPA, a worsening of asthma alone, however, is usually insufficient to diagnose a relapse of EGPA.
Many patients are prescribed chronic low dose of prednisone to help with a control of asthma rather than to treat vasculitis or parenchymal eosinophilic inflammation. When both upper and lower respiratory symptoms recur upon prednisone tapering, it is often difficult to decide whether vasculitis itself is active. Blood eosinophil counts are reduced by even low doses of glucocorticoids and are therefore also difficult to interpret in clinical practice. Conversely, elevated eosinophil counts may occur without return of vasculitis.
We studied a set of potential biomarkers (Eotaxin-3, TARC/CCL17, IgG4, and IgG4/IgG ratio) for disease relapse in a prospective, well-defined EGPA cohort. All biomarkers have been previously identified in newly- diagnosed patients with EGPA. TARC/CCL17, a chemokine responsible for Th2 cell recruitment, is elevated in the serum of active patients with EGPA correlating with disease activity [8]. Further, eotaxin-3 is an eosinophil-attracting chemokine elevated in EGPA but not in other eosinophilic or vasculitic disorders [14]. IgG4 is the least abundant IgG subclass. The function of IgG4 is currently unclear but it is associated with a Th2 immune response [9]. IgG4 is highly elevated in active EGPA at disease presentation and decreases during remission. However, prior studies of IgG4 in EGPA included patients who were newly diagnosed and most patients were not on immunosuppressive therapy at the time of testing. Therefore, we studied a real-life cohort of established patients with EGPA focusing on relapsing disease.
In this cohort, most disease flares were mild in the observational period and we could not find any significant differences in routine serologic markers such as ESR, CRP, or eosinophil count. These findings have been corroborated in a more detailed analysis of the same VCRC cohort using data from 141 patients with EGPA with longitudinal follow-up [15]. Organ manifestations of relapsing patients were, however, typical for EGPA with a predominance of constitutional and respiratory symptoms, ENT manifestations and neuropathic pain. Levels of eotaxin-3, TARC/CCL17, IgG4, and IgG4/IgG ratio all failed to discriminate reliably between active and inactive patients. Prednisone clearly modulated serum levels of eotaxin-3 and eosinophils independent of disease activity. Overall, the frequent use of prednisone in EGPA argues against a significant role for these biomarkers in follow-up clinical care. Whether immunosuppressive agents also influence levels of the measured biomarkers is unclear. In our cohort, no statistically significant association was observed in the multivariate analysis; however, the number of patients on immunosuppressive agents was small and several agents with potentially different impact on biomarkers were used. Larger studies are required to investigate the possible association between immunosuppressive drugs and biomarkers in EGPA.
Our study has some important strengths. The study subjects were part of a standardized, prospective, longitudinal data collection with an extensive dataset. In addition, all patients were recruited in expert centers and assessed by clinicians familiar with this rare disease and its clinical manifestations.
Our study has also some limitations: The sample size of our cohort was small compared to biomarker studies in other rheumatic diseases. However, given the rarity of EGPA, it is challenging to assemble larger cohorts for such a project. Most relapses were mild lacking serious clinical manifestations (such as cardiovascular, gastrointestinal and central nervous system involvement), and we cannot exclude the possibility that some patients had their prednisone dose increased before serum sampling. Yet, both of these facts reflect the clinical reality of the disease and its management. Moreover, most patients were already on glucocorticoids at the initiation of study, and detailed information about doses of prednisone and immunosuppressive agents used prior to enrollment were limited, making the analysis of associations between biomarkers and these factors difficult.
In summary, defining biomarkers for relapsing disease in EGPA remains difficult. Novel biomarkers tested in this study could not reliably discriminate active and inactive disease. The potential impact of glucocorticoids and other immunosuppressive drugs with available and potential biomarkers remains a major challenge in the study of biomarkers for EGPA.
Author Contributions
Conceived and designed the experiments: CD PM PAM JZ. Performed the experiments: JZ. Analyzed the data: CD BO PM DC SC GH NK CK CL KM PS US SY PAM JZ. Contributed reagents/materials/analysis tools: PM DC SC GH NK CK CL KM PS US SY PAM. Wrote the paper: CD BO PM DC SC GH NK CK CL KM PS US SY PAM JZ. Statistical analysis: CD.
References
- 1. Churg J, Strauss L (1951) Allergic granulomatosis, allergic angiitis, and periarteritis nodosa. Am J Pathol 27: 277–301. pmid:14819261
- 2. Moosig F, Bremer JP, Hellmich B, Holle JU, Holl-Ulrich K, Laudien M, et al. (2013) A vasculitis centre based management strategy leads to improved outcome in eosinophilic granulomatosis and polyangiitis (Churg-Strauss, EGPA): monocentric experiences in 150 patients. Ann Rheum Dis 72: 1011–7. pmid:22887848
- 3. Neumann T, Manger B, Schmid M, Kroegel C, Hansch A, Kaiser WA, et al. (2009) Cardiac involvement in Churg-Strauss syndrome: impact of endomyocarditis. Medicine (Baltimore) 88: 236–43. pmid:19593229
- 4. Vaglio A, Buzio C, Zwerina J (2013) Eosinophilic granulomatosis with polyangiitis (Churg-Strauss): state of the art. Allergy 68: 261–73. pmid:23330816
- 5. Mahr A, Moosig F, Neumann T, Szczeklik W, Taille C, Vaglio A, et al. (2014) Eosinophilic granulomatosis with polyangiitis (Churg-Strauss): evolutions in classification, etiopathogenesis, assessment and management. Curr Opin Rheumatol 26: 16–23. pmid:24257370
- 6. Moosig F, Gross WL, Herrmann K, Bremer JP, Hellmich B (2011) Targeting interleukin-5 in refractory and relapsing Churg-Strauss syndrome. Ann Intern Med 155: 341–3. pmid:21893636
- 7. Comarmond C, Pagnoux C, Khellaf M, Cordier JF, Hamidou M, Viallard JF, et al. (2013) Eosinophilic granulomatosis with polyangiitis (Churg-Strauss): clinical characteristics and long-term followup of the 383 patients enrolled in the French Vasculitis Study Group cohort. Arthritis Rheum 65: 270–81. pmid:23044708
- 8. Dallos T, Heiland GR, Strehl J, Karonitsch T, Gross WL, Moosig F, et al. (2010) CCL17/thymus and activation-related chemokine in Churg-Strauss syndrome. Arthritis Rheum 62: 3496–503. pmid:20669282
- 9. Vaglio A, Strehl JD, Manger B, Maritati F, Alberici F, Beyer C, et al. (2012) IgG4 immune response in Churg-Strauss syndrome. Ann Rheum Dis 71: 390–3. pmid:22121132
- 10. Polzer K, Karonitsch T, Neumann T, Eger G, Haberler C, Soleiman A, et al. (2008) Eotaxin-3 is involved in Churg-Strauss syndrome—a serum marker closely correlating with disease activity. Rheumatology (Oxford) 47: 804–8. pmid:18397958
- 11. Masi AT, Hunder GG, Lie JT, Michel BA, Bloch DA, Arend WP, et al. (1990) The American College of Rheumatology 1990 criteria for the classification of Churg-Strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum 33: 1094–100. pmid:2202307
- 12. Stone JH, Hoffman GS, Merkel PA, Min YI, Uhlfelder ML, Hellmann DB, et al. (2001) A disease-specific activity index for Wegener's granulomatosis: modification of the Birmingham Vasculitis Activity Score. International Network for the Study of the Systemic Vasculitides (INSSYS). Arthritis Rheum. 44: 912–20. pmid:11318006
- 13. Luqmani RA, Bacon PA, Moots RJ, Janssen BA, Pall A, Emery P, et al. (1994) Birmingham Vasculitis Activity Score (BVAS) in systemic necrotizing vasculitis. QJM 87: 671–8. pmid:7820541
- 14. Zwerina J, Bach C, Martorana D, Jatzwauk M, Hegasy G, Moosig F, et al. (2011) Eotaxin-3 in Churg-Strauss syndrome: a clinical and immunogenetic study. Rheumatology (Oxford) 50: 1823–7. pmid:21266446
- 15. Grayson PC, Monach PA, Pagnoux C, Cuthbertson D, Carette S, Hoffman GS, et al. (2014) Value of commonly measured laboratory tests as biomarkers of disease activity and predictors of relapse in eosinophilic granulomatosis with polyangiitis. Rheumatology (Oxford) keu 427 pmid:25406357