UvA-DARE (Digital Academic Repository)

Towards new measures of inflammation in spondyloarthritis

Turina, M.C.

Publication date 2016 Document Version Final published version

Link to publication

Citation for published version (APA): Turina, M. C. (2016). Towards new measures of inflammation in spondyloarthritis.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:26 Sep 2021 Towards new measures of inflammation in spondyloarthritis Maureen Cindy Turina

TOWARDS NEW MEASURES OF INFLAMMATION IN SPONDYLOARTHRITIS

UITNODIGING

Voor het bijwonen van de openbare verdediging van het proefschrift van

Maureen Turina

op vrijdag 1 april 2016 om 12:00 uur

in de Agnietenkapel Oudezijds Voorburgwal 231, Amsterdam

Receptie ter plaatse na afloop van de promotie

Towards new measures of Paranimfen Jacky Paramarta: inflammation in spondyloarthritis j.e[email protected] Maureen Turina Talia Latuhihin: t[email protected] TOWARDS NEW MEASURES OF INFLAMMATION IN SPONDYLOARTHRITIS

Maureen Cindy Turina ISBN: 978-94-6233-234-8 Maureen Turina was supported by a fellowship of Janssen.

Printing of this thesis was financially supported by ABBVIE B.V., AMC-UvA, the Dutch Arthritis Foundation (het Reumafonds), Janssen-Cilag B.V., Pfizer B.V., and UCB Pharma B.V., which is gratefully acknowledged.

Cover design and thesis layout: Midas Mentink (www.thesisexpert.nl) Printing: Gildeprint – Enschede (www.gildeprint.nl) Copyright © 2015 by M.C. Turina

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without permission of the author. TOWARDS NEW MEASURES OF INFLAMMATION IN SPONDYLOARTHRITIS

ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op vrijdag 1 april 2016, te 12:00 uur

door Maureen Cindy Turina geboren te Rotterdam PROMOTIECOMMISSIE:

Promotores: Prof. dr. D.L.P. Baeten Universiteit van Amsterdam Prof. dr. R.B.M. Landewé Universiteit van Amsterdam

Overige leden: Prof. dr. D. van Schaardenburg Universiteit van Amsterdam Prof. dr. P.I. Spuls Universiteit van Amsterdam Prof. dr. G.R.A.M. D’Haens Universiteit van Amsterdam Prof. dr. P.L.C.M. van Riel Radboud Universiteit Nijmegen Dr. F.A. van Gaalen Universiteit Leiden

Faculteit der Geneeskunde

TABLE OF CONTENTS

Chapter 1 General introduction 9

Chapter 2 Serum biomarkers in spondyloarthritis-related diseases: 19 lessons from psoriasis and inflammatory bowel diseases

Chapter 3 Serum inflammatory biomarkers fail to identify early axial 43 spondyloartritis: results from the SpondyloArthritis Caught Early (SPACE)-cohort

Chapter 4 Clinical and imaging signs of spondyloarthritis in first-degree 55 relatives of HLA-B27 positive ankylosing spondylitis patients: the pre-spondyloarthritis (Pre-SpA) cohort

Chapter 5 Calprotectin (S100A8/9) as serum biomarker for clinical 75 response in proof-of-concept trials in axial and peripheral spondyloarthritis

Chapter 6 A psychometric analysis of outcome measures in peripheral 91 spondyloarthritis

Chapter 7 Calprotectin serum level is an independent marker for 107 radiographic spinal progression in axial spondyloarthritis

Chapter 8 General discussion and summary 113

Appendices Nederlandse samenvatting 135 Dankwoord 143 PhD portfolio 147 Curriculum vitae 151 List of publications 153

General introduction 1 SPONDYLOARTHRITIS

Spondyloarthritis (SpA) is a chronic inflammatory disease affecting the axial skeleton, the peripheral joints, and extra-articular organs such as eye, gut, and skin. An important symptom of axial SpA is inflammatory back pain with pain at night and morning stiffness. Spinal inflammation can be accompanied by excessive new bone formation leading to fusion of vertebral bodies and spinal ankylosis. The prototypical and best known form of axial SpA is ankylosing spondylitis (AS). Peripheral SpA may manifest as peripheral arthritis, enthesitis, and/or dactylitis. Peripheral arthritis is characterized by pain, swelling, redness, and stiffness of the joints; the arthritis is mainly of the oligoarticular (1-5 joints) type, with asymmetrical distribution, preferably affecting the large joints of the lower limbs. Peripheral arthritis can be associated with both bone destruction and new bone formation, as described in psoriatic arthritis, being the prototypical form of peripheral SpA. Enthesitis often affects the fascia plantaris and the Achilles tendon. Dactylitis is swelling of an entire digit of the hands (sausage finger) or feet (sausage toe). Extra-articular manifestations include acute anterior uveitis, inflammatory bowel disease (Crohn’s disease and ulcerative colitis) and skin psoriasis. The inflammatory musculoskeletal symptoms of SpA are often effectively treated with nonsteroidal anti-inflammatory drugs (NSAIDs) and, if refractory to NSAIDs, tumor necrosis factor (TNF) inhibitors. TNF inhibitors do inhibit bone destruction in psoriatic arthritis. In contrast, it is still unclear if TNF-inhibitors inhibit new bone formation as seen in both axial and peripheral SpA.

Unmet needs in the management of SpA Although the management of SpA has improved remarkably in the past years, mainly due to TNF inhibitors, several unmet needs still remain.

A first unmet need is the time required to diagnose SpA. For axial disease, for example, until recently, it took ten years on average from the start of signs and symptoms to diagnose SpA.1,2 This delay has decreased significantly by optimizing the referral strategies from the general practitioner to the rheumatologist,3–5 and by introducing the magnetic resonance imaging (MRI) that is able to depict inflammation of the sacroiliiac joints (SI-joints). Still, however, the average delay between first symptoms and diagnosis of axial SpA remains 5-10 years,6,7 reflecting a long period of uncontrolled signs and symptoms. Comparable data for peripheral SpA are not available. Therefore, we need to study the earlier (yet undiagnosed) phases of clinically established SpA. In addition, the study of the subclinical phases of SpA (i.e. individuals without the symptoms of SpA but already with certain abnormalities specific for SpA (imaging, laboratory tests)) may further help to reduce the unwarranted diagnostic delay. Two main arguments explain why any delay in diagnosing SpA is unwarranted: 1) effective treatment is available to immediately improve signs and symptoms of the disease without further delay,8–14 and 2) treatment during the early phases of the disease may

10 GENERAL INTRODUCTION better halt new bone formation and the progression thereof than late-onset treatment after significant diagnostic delay (window of opportunity in SpA).15–17 1 A second unmet need is the lack of alternative therapeutic options beyond TNF inhibitors. Despite their major efficacy, TNF inhibitors have several limitations: 1) approximately 40% of the patients do not respond well to TNF inhibitor treatment;8,10–13 2) contra-indications for TNF inhibitors usage and/or side-effects occurring during TNF inhibitor-use limit the clinical potential of this class of treatments; 3) while TNF blockade can halt joint destruction, it likely does not halt new bone formation;15–17 and 4) long-term drug-free remission can still not be achieved.18,19 Clinical studies of novel drugs therefore remain necessary but are rapidly becoming ethically challenging: sufficiently large, long-term placebo controlled trials are ethically disputable in the era of TNF blockade, and difficult to enrol. Therefore, there is a need for alternative strategies to quickly provide go or no-go signals on whether a new drug may show efficacy at the group level in small-scale short-term proof of concept clinical trials (PoCs) before proceeding with larger phase-2b/3 randomized controlled trials.

A third and partially related unmet need is to find valid outcome measures that adequately reflect treatment responses in peripheral disease. In axial SpA, well validated outcome measures such as the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI)20 and the Ankylosing Spondylitis Diseases Activity Score (ASDAS)21 exist and are widely used in clinical studies, including therapeutic trials. In peripheral SpA, such outcome measures are lacking, with the exception of psoriatic arthritis: the outcome measures for psoriatic arthritis were either derived from measures originally developed for rheumatoid arthritis (e.g. American College of Rheumatology criteria, ACR) or were newly developed specifically for psoriatic arthritis, including Psoriatic Arthritis Response Criteria (PsARC). In peripheral SpA, Peripheral SpA Response Criteria (PSpARC) is the only set of response criteria developed for peripheral SpA. The discriminatory capacity of these outcome measures has not been properly validated for peripheral SpA. Thus, patient-reported outcome measures or markers determining the efficacy of treatments at the group level are needed in peripheral SpA.

A fourth important unmet need is inhibition of pathological (inappropriate) new bone formation.15–17 New bone formation and the resulting joint ankylosis may impair physical function and thereby quality of life in affected patients. The challenge here is notonly to develop novel treatment that, in contrast with current treatments, will halt new bone formation at the group level, but rather to identify those patients that may benefit most of these treatments. Indeed, not all patients with SpA will develop syndesmophytes and ankylosis. Identifying the patients at highest risk of rapid structural progression and treating them accordingly early may be as important to improve outcome in SpA as developing novel treatments.

CHAPTER 1 11 Biomarkers One approach that could help to address the aforementioned challenges is the identification and validation of biomarkers in SpA. A biomarker is defined as a “characteristic that can be objectively measured and evaluated as an indicator of a normal biologic process, a pathophysiologic process, or a pharmacologic response to a therapeutic intervention”.22 A biomarker should ideally be sensitive, specific and reproducible, and obtaining the biomarker should preferably be by a non-invasive procedure. Taking these criteria into account, biomarkers can have an important role in fulfilling the unmet needs in the process of diagnosis, treatment response, and clinical outcome. Examples of valuable biomarkers in other fields are: 1) anti-citrullinated protein antibodies (ACPA) in rheumatoid arthritis. This marker is highly specific for diagnosis (specificity of 95%), has good reproducibility, and is easily accessible as it is determined in the serum. Approximately 30% of patients with arthralgia who are ACPA positive will develop RA within one year,23–25 and therefore these antibodies may identify high-risk patients. 2) The human epidermal growth factor receptor 2 (HER2)-status can predict responses to systemic treatment in breast cancer, including trastuzumab (Herceptin®, a monoclonal IgG1 antibody against HER2) or chemotherapeutic agents, including anthracycline- and taxane-containing regimens.26–29 3) CD4+ cell count and viral load in human immunodeficiency virus (HIV), both of which are good prognostic biomarkers which can predict the outcomes such as acquired immune deficiency syndrome (AIDS) and death.30–32

Also in the field of SpA, biomarkers may play an increasingly important role in addressing major clinical challenges such as early diagnosis, prediction and measurement of treatment responses, and prognosis.

AIMS AND OUTLINE OF THIS THESIS

In this thesis we aimed to study the value of serum biomarkers to address the aforementioned major unmet needs in SpA. The main aims were as follows:

1. Diagnosis: to identify (bio-)markers (clinical, imaging and laboratory markers) which can help diagnosing SpA in the early phases of the clinically established disease or already during the subclinical phases of the disease; 2. Treatment response: to identify serum biomarkers for treatment responses in small- scale, short-term, PoC trials, and to validate outcome measures for peripheral SpA; 3. Treatment response: to identify outcome measures for peripheral SpA; 4. Outcome: to identify biomarkers which are predictive for the progression of structural damage in the spine in axial SpA.

12 GENERAL INTRODUCTION Early diagnosis Concerning the first unmet need, i.e. markers for early SpA diagnosis, there are no serum 1 markers known in literature that are able to diagnose a patient with SpA at the early phase of the disease. In order to address this unmet need, we used three different approaches, with the ultimate goal to find early diagnostic biomarkers. In chapter 2, we performed a literature search in the field of psoriasis and inflammatory bowel disease (IBD), both disorders which are clinically, genetically, and pathophysiologically related to SpA, in order to find potential new serum biomarkers of interest for SpA. In chapter 3, we determined the value of serum biomarkers to discriminate SpA from non-SpA in a cohort of patients with early inflammatory back pain. First, we evaluated the discriminative and diagnostic value of C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and calprotectin, all of which are known to be elevated in active axial SpA.33,34 Second, we explored whether the informative serum biomarkers in psoriasis and IBD (as mentioned in chapter 2), namely human beta defensin-2 (hBD-2) and lipocalin-2 (LCN-2), are elevated in the cohort of active axial SpA versus healthy controls. Third, we also checked if IL-27, an IL-12 family cytokine, is elevated in full blown AS when compared to healthy controls as indicated in a recent report.

In chapter 4, we performed a systematic screening in seemingly healthy first-degree relatives (FDRs) of HLA-B27 positive AS patients, since FDRs are at a high risk of developing SpA. In this early phase of SpA, we attempt to identify pre- or subclinical biomarkers, and imaging alterations in these FDRs without SpA symptoms.

Treatment response Concerning the second unmet need, i.e. markers for treatment response, it is important to have objective markers which give quick go or no-go signals on whether a new treatment is effective at the group level in PoC trials. Inchapter 5 we determined which serum biomarkers were able to reflect quickly and reproducibly clinical responses at the group level in PoC trials in SpA. These markers may be useful to predict a clinical response in larger and/or longer term trials. Additionally, in chapter 6 we performed an analysis in two independent randomized controlled trials to determine which clinical disease outcome measures do best reflect disease activity- and treatment responses in peripheral SpA.

Prediction of structural damage Concerning the fourth unmet need, i.e. halting new bone formation, it is important to predict which patients will develop rapidly progressive new bone formation of the spine. Smoking, HLA-B27 positive status, male gender, and a high disease activity have all been reported to be associated with more rapid progression of structural damage in AS.35–37 Also serum biomarkers such as CRP,38,39 matrix metalloproteinase (MMP-3)40, dickkopf-1 (Dkk-1),36,41,42 visfatin and leptin,43–45 and vascular endothelial growth factor (VEGF)46 have

CHAPTER 1 13 been reported to be associated with structural damage of the spine but none of these aforementioned biomarkers could reproducibly predict progression of structural damage in individual patients. Inchapter 7, we determined whether serum levels of calprotectin are an independent predictor for new bone formation in axial SpA.

Finally, chapter 8 provides a general discussion and summary of the findings of this thesis.

14 GENERAL INTRODUCTION REFERENCES 1 1. Feldtkeller E, Khan MA, van der Heijde D, et al. Age at 2006;54(7):2136–46. disease onset and diagnosis delay in HLA-B27 negative 14. Landewé R, Braun J, Deodhar a, et al. Efficacy of vs. positive patients with ankylosing spondylitis. certolizumab pegol on signs and symptoms of axial Rheumatol Int 2003;23(2):61–6. spondyloarthritis including ankylosing spondylitis: 2. Bakland G, Nossent HC, Gran JT. Incidence and 24-week results of a double-blind randomised prevalence of ankylosing spondylitis in Northern placebo-controlled Phase 3 study. Ann Rheum Dis Norway. Arthritis Rheum 2005;53(6):850–5. 2014;73(1):39–47. 3. Brandt HC, Spiller I, Song I-H, et al. Performance of 15. Van der Heijde D, Landewé R, Einstein S, et al. referral recommendations in patients with chronic back Radiographic progression of ankylosing spondylitis pain and suspected axial spondyloarthritis. Ann Rheum after up to two years of treatment with etanercept. Dis 2007;66(11):1479–84. Arthritis Rheum 2008;58(5):1324–31. 4. Poddubnyy D, Vahldiek J, Spiller I, et al. Evaluation 16. Van der Heijde D, Landewé R, Baraliakos X, et al. of 2 screening strategies for early identification of Radiographic findings following two years of infliximab patients with axial spondyloarthritis in primary care. J therapy in patients with ankylosing spondylitis.Arthritis Rheumatol 2011;38(11):2452–60. Rheum 2008;58(10):3063–70. 5. Sieper J, Srinivasan S, Zamani O, et al. Comparison 17. Van der Heijde D, Salonen D, Weissman BN, et al. of two referral strategies for diagnosis of axial Assessment of radiographic progression in the spines spondyloarthritis: the Recognising and Diagnosing of patients with ankylosing spondylitis treated with Ankylosing Spondylitis Reliably (RADAR) study. Ann adalimumab for up to 2 years. Arthritis Res Ther Rheum Dis 2013;72(10):1621–7. 2009;11(4):R127. 6. Salvadorini, F. Bandinelli, A. Delle Sedie, L. Riente, A. 18. Baraliakos X, Listing J, Brandt J, et al. Clinical response Candelieri, S. Generini, N. Possemato, S. Bombardieri to discontinuation of anti-TNF therapy in patients with MM-C. Ankylosing spondylitis: how diagnostic and ankylosing spondylitis after 3 years of continuous therapeutic delay have changed over the last six treatment with infliximab. Arthritis Res Ther decades. Clin Exp Rheumatol 2012;30(4):561–5. 2005;7(3):R439–44. 7. Sorensen J, Hetland ML. Decreases in diagnostic delay 19. Paramarta JE, Heijda TF, Baeten DL. Fast relapse upon are supported by sensitivity analyses. Ann Rheum Dis discontinuation of tumour necrosis factor blocking 2014;73(7):e45–e45. therapy in patients with peripheral spondyloarthritis. 8. Van Den Bosch F, Kruithof E, Baeten D, et al. Randomized Ann Rheum Dis 2013;72(9):1581–2. double-blind comparison of chimeric monoclonal 20. Garrett S, Jenkinson T, Kennedy LG, Whitelock H, antibody to tumor necrosis factor alpha (infliximab) Gaisford P, Calin A. A new approach to defining disease versus placebo in active spondylarthropathy. Arthritis status in ankylosing spondylitis: the Bath Ankylosing Rheum 2002;46(3):755–65. Spondylitis Disease Activity Index. Rheumatology 9. Braun J, Brandt J, Listing J, et al. Treatment of 1994;21(12):2286–91. active ankylosing spondylitis with infliximab: a 21. Lukas C, Landewé R, Sieper J, et al. Development of randomised controlled multicentre trial. Lancet an ASAS-endorsed disease activity score (ASDAS) in 2002;359(9313):1187–93. patients with ankylosing spondylitis. Ann Rheum Dis 10. Mease PJ, Goffe BS, Metz J, VanderStoep A, Finck B, 2009;68(1):18–24. Burge DJ. Etanercept in the treatment of psoriatic 22. Gruttola VG De, Clax P, Demets DL, et al. Considerations arthritis and psoriasis: a randomised trial. Lancet in the Evaluation of Surrogate Endpoints in Clinical 2000;356(9227):385–90. Trials: Summary of a National Institutes of Health 11. Paramarta JE, De Rycke L, Heijda TF, et al. Efficacy and Workshop. Control Clin Trials 2001;1998(2001):485– safety of adalimumab for the treatment of peripheral 502. arthritis in spondyloarthritis patients without 23. Bos WH, Wolbink GJ, Boers M, et al. Arthritis ankylosing spondylitis or psoriatic arthritis. Ann Rheum development in patients with arthralgia is strongly Dis 2013;72(11):1793–9. associated with anti-citrullinated protein antibody 12. Sieper J, van der Heijde D, Dougados M, et al. Efficacy status: a prospective cohort study. Ann Rheum Dis and safety of adalimumab in patients with non- 2010;69(3):490–4. radiographic axial spondyloarthritis: results of a 24. Nielen MMJ, van Schaardenburg D, Reesink HW, et randomised placebo-controlled trial (ABILITY-1). Ann al. Specific autoantibodies precede the symptoms of Rheum Dis 2013;72(6):815–22. rheumatoid arthritis: a study of serial measurements in 13. Van der Heijde D, Kivitz A, Schiff MH, et al. Efficacy blood donors. Arthritis Rheum 2004;50(2):380–6. and safety of adalimumab in patients with ankylosing 25. Nishimura K, Sugiyama D, Kogata Y, Tsuji G, Nakazawa spondylitis: results of a multicenter, randomized, T. Annals of Internal Medicine Review Meta-analysis : double-blind, placebo-controlled trial. Arthritis Rheum Diagnostic Accuracy of Anti–Cyclic Citrullinated Peptide

CHAPTER 1 15 Antibody and Rheumatoid Factor for Rheumatoid 40. Yang C, Gu J, Rihl M, et al. Serum levels of matrix Arthritis. Ann Intern Med 2015;797. metalloproteinase 3 and macrophage colony- 26. Joensuu H, Kellokumpu-Lehtinen P-L, Bono P, stimulating factor 1 correlate with disease activity in et al. Adjuvant docetaxel or vinorelbine with or ankylosing spondylitis.Arthritis Rheum 2004;51(5):691– without trastuzumab for breast cancer. N Engl J Med 9. 2006;354(8):809–20. 41. Klingberg E, Nurkkala M, Carlsten H, Forsblad-d’Elia 27. Konecny GE, Thomssen C, Lück HJ, et al. Her-2/neu H. Biomarkers of bone metabolism in ankylosing gene amplification and response to paclitaxel in spondylitis in relation to osteoproliferation and patients with metastatic breast cancer. J Natl Cancer osteoporosis. J Rheumatol 2014;41(7):1349–56. Inst 2004;96(15):1141–51. 42. Korkosz M, Jerzy G, Leszczy P, Pawlak-bu K, Jeka 28. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus S, Kucharska E. High disease activity in ankylosing Adjuvant Chemotherapy for Operable HER2-Positive spondylitis is associated with increased serum Breast Cancer. N Engl J Med 2005;353:1673–84. sclerostin level and decreased wingless protein-3a signaling but is not linked with greater structural 29. Pritchard KI, Shepherd LE, O’Malley FP, et al. HER2 damage. BMC Musculoskelet Disord. 2013; 14:99. and responsiveness of breast cancer to adjuvant chemotherapy. N Engl J Med 2006;354(20):2103–11. 43. Kim K-J, Kim J-Y, Park S-J, et al. Serum leptin levels are associated with the presence of syndesmophytes 30. Giorgi J V, Lyles RH, Matud JL, et al. Predictive value of in male patients with ankylosing spondylitis. Clin immunologic and virologic markers after long or short Rheumatol 2012;31(8):1231–8. duration of HIV-1 infection. J Acquir Immune Defic Syndr 2002;29:346–55. 44. Syrbe U, Callhoff J, Conrad K, et al. Serum adipokine levels in patients with ankylosing spondylitis and their 31. Mellors JW, Munoz A, Giorgi J V, et al. Plasma Viral Load relationship to clinical parameters and radiographic and CD4 + Lymphocytes as Prognostic Markers of HIV-1 spinal progression. Arthritis Rheumatol (Hoboken, NJ) Ann Intern Med Infection. 1997;126:946–54. 2015;67(3):678–85. 32. Bruisten S, Frissen PHJ, Swieten PVAN, et al. Prospective 45. Genre F, López-Mejías R, Miranda-Filloy J a, et al. Longitudinal Analysis of Viral Load and Surrogate Adipokines, biomarkers of endothelial activation, Markers in Relation to Clinical Progression in HIV and metabolic syndrome in patients with ankylosing AIDS Res Hum Retroviruses Type 1-Infected Persons. spondylitis. Biomed Res Int 2014;2014:860651. 1997;13(4):327–35. 46. Poddubnyy D, Conrad K, Haibel H, et al. Elevated serum 33. Kane D, Roth J, Frosch M, Vogl T, Bresnihan B, FitzGerald level of the vascular endothelial growth factor predicts O. Increased perivascular synovial membrane radiographic spinal progression in patients with axial expression of myeloid-related proteins in psoriatic spondyloarthritis. Ann Rheum Dis 2014;73(12):2137– Arthritis Rheum arthritis. 2003;48(6):1676–85. 43. 34. De Rycke L, Baeten D, Foell D, et al. Differential expression and response to anti-TNFalpha treatment of infiltrating versus resident tissue macrophage subsets in autoimmune arthritis. J Pathol 2005;206(1):17–27. 35. Ramiro S, Stolwijk C, van Tubergen A, et al. Evolution of radiographic damage in ankylosing spondylitis: a 12 year prospective follow-up of the OASIS study. Ann Rheum Dis 2015;74(1):52–9. 36. Heiland GR, Appel H, Poddubnyy D, et al. High level of functional dickkopf-1 predicts protection from syndesmophyte formation in patients with ankylosing spondylitis. Ann Rheum Dis 2012;71(4):572–4. 37. Ramiro S, van der Heijde D, van Tubergen A, et al. Higher disease activity leads to more structural damage in the spine in ankylosing spondylitis: 12-year longitudinal data from the OASIS cohort. Ann Rheum Dis 2014;73(8):1455–61. 38. Poddubnyy D, Rudwaleit M, Haibel H, et al. Rates and predictors of radiographic sacroiliitis progression over 2 years in patients with axial spondyloarthritis. Ann Rheum Dis 2011;70(8):1369–74. 39. Pedersen SJ, Sørensen IJ, Lambert RGW, et al. Radiographic progression is associated with resolution of systemic inflammation in patients with axial spondylarthritis treated with tumor necrosis factor α inhibitors. Arthritis Rheum 2011;63(12):3789–800.

16 GENERAL INTRODUCTION 1

CHAPTER 1 17

Serum biomarkers in spondyloarthritis- related diseases: lessons from psoriasis and inflammatory bowel diseases

Maureen C. Turina, Robert Landewé, Dominique L. Baeten

Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and immunology Center, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands 2

Submitted for publication SUMMARY

Early diagnosis, monitoring of disease activity, prediction of treatment response, and structural outcome remain major challenges in spondyloarthritis (SpA). Biomarkers could play a role in addressing these challenges, but in SpA there is a lack of suitable biomarkers. As SpA is clinically and pathophysiologically closely related to psoriasis and inflammatory bowel disease (IBD), we reviewed the value of serum biomarkers in these conditions with the aim to find potential candidates for assessing SpA. Candidates of interest were antimicrobial peptides, including human beta defensin-2 (hBD-2) and lipocalin-2 (LCN-2), and class-1 MHC molecule beta2-microglobulin. Since these biomarkers are relevant in psoriasis and/or IBD from a pathophysiological point of view, and may play a role in the pathogenesis of SpA, we recommend further exploration of their value as biomarker in the diagnosis and prognosis of SpA.

20 SERUM BIOMARKERS IN PSORIASIS AND IBD UNMET NEEDS IN SPONDYLOARTHRITIS

Spondyloarthritis (SpA) is an immune-mediated inflammatory disease affecting the spine, the peripheral joints and extra-articular tissues such as the skin, the gut, and theeye.1 2 The pathology of SpA is characterized by a combination of chronic tissue inflammation, destruction of cartilage and bone, and pathological new bone formation, that maylead to ankylosis. Despite major advances in the field, such as the use of magnetic resonance imaging (MRI) to detect spinal inflammation and to find appropriate patients for treatment with TNF blockers,2–8 several important challenges remain to be addressed.

First, there is a huge diagnostic delay of approximately 5-10 years between the onset of SpA symptoms and the diagnosis.9–11 This delay is due to the fact that the early clinical signs and symptoms are often non-specific, that pathognomonic features of SpA, such as radiologic sacroiliitis, only appear late in the disease course, and that well-known biological markers such as HLA-B27 lack specificity (many healthy individuals carry HLA-B27) and therefore predictive value in the diagnostic context. Second, good biological predictors for treatment response are lacking. Currently, the only informative marker is C-reactive protein (CRP). Nevertheless, CRP as a marker for treatment response falls short because only one third of the patients have elevated CRP levels prior to treatment. Third, the progression rate of new bone formation as observed on consecutive X-rays of the lumbar and cervical spine is very slow and besides only occurs in a relatively small group of patients. Currently, markers predicting which patients will develop significant new bone formation are lacking.

Overall, an early diagnosis, a proper prediction of structural damage, as well as prediction of treatment response, all in individual patients, remain major challenges in SpA. Biomarkers could be of value in addressing these challenges.

BIOMARKERS IN SPONDYLOARTHRITIS

A biomarker is a “characteristic that can be objectively measured and evaluated as an indicator of a normal biologic process, a pathophysiologic process, or a pharmacologic response to a therapeutic intervention”.12 A biomarker should be sensitive, specific, reproducible, and deriving the biomarker from a patient should preferably be a non-invasive procedure. Taking these criteria into account, biomarkers can have an important role to fulfill the previously defined unmet needs in the process of diagnosis, treatment response and prognosis. Several biomarkers have already been studied in SpA. Common markers of systemic inflammation such as CRP and erythrocyte sedimentation rate (ESR) have been used for a long time in the clinical care of SpA patients. Accordingly, elevated CRP is included as a criterion in the ASAS classification criteria for axial SpA criteria but not for ASAS peripheral SpA criteria.2,13

CHAPTER 2 21 When elevated, CRP levels may correlate with disease activity measures and decrease upon effective treatment.14–17 Moreover, CRP appears to be a predictor for radiographic axial progression.18 However, CRP and ESR are only elevated in 30 to 50% of the SpA patients with clinical signs and symptoms of active disease and inflammation14–17 and, obviously, are not specific for SpA. The limited sensitivity and specificity explains why CRP and ESR can be useful biomarkers at the group level (such as in clinical trials) but lack sufficient predictive value to be used in individual patients.

Other serum markers of inflammation that have been studied in SpA are Interleukin-6 (IL- 6) and calprotectin. IL-6 is the main driver of CRP and, similar to CRP, serum IL-6 levels are significantly increased in active SpA and decrease upon clinical response after tumour necrosis factor (TNF) blockade.19,20 The calcium-binding protein calprotectin, a heterodimer of S100A8 and S100A9, is expressed and secreted during monocyte infiltration into inflamed tissues, including macrophage infiltration in synovial tissue of SpA patients.21,22 Calprotectin serum levels are elevated in SpA, correlate moderately well (r=0.40-0.60) with disease activity, and decrease upon TNF blockade.21–23 Moreover, calprotectin serum levels is an independent predictor for radiological axial progression at the group level.24 Two other biomarkers that were studied in SpA are more related to the pathological processes than to inflammation as such. Vascular endothelial growth factor (VEGF), a growth factor involved in neo-angiogenesis, may be involved in the pronounced hypervascularity in SpA synovitis.25,26 Serum VEGF levels are significantly elevated in SpA versus healthy controls and significantly albeit moderately at best correlate (r=0.22-0.44) with disease activity, but levels do not decrease after short-term treatment with TNF blockers.20,27 Whereas an original report indicated that serum VEGF levels were predictive of new bone formation,28 these results could not be confirmed by others.29

Matrix metalloproteinase-3 (MMP-3) is an enzyme involved in tissue remodelling and cartilage damage. MMP-3 expression was elevated in synovial tissue, synovial fluid and serum in SpA and this expression correlated well with disease activity30–34 and normalized upon effective treatment.30

With regard to predicting new bone formation in ankylosing spondylitis (AS), the prototype of axial SpA, most studies have focused on molecules involved in the regulation of osteoblast activity, including the wnt-pathway, which is involved in the process of new bone formation. Dickkopf-1 (DKK1) and sclerostin are inhibitors of this pathway, and Diarra et al.35 found that serum DKK1 levels were decreased in AS patients versus healthy controls or rheumatoid arthritis patients and low levels were associated to radiographic axial progression as expressed by syndesmophytes formation. These results were confirmed by others.36–38 For sclerostin, discordant results were seen regarding the relationship with radiographic axial spinal progression.36,39–42

Finally, several studies have screened for disease-specific autoantibodies as potential

22 SERUM BIOMARKERS IN PSORIASIS AND IBD diagnostic biomarker in SpA. Unfortunately, most studies did not find disease-specific autoantibodies and the few that were reported could not be independently confirmed.43–45 Recently, one group found elevated levels of IgG autoantibodies CD74 specific for HLA class II-associated invariant chain peptide (CLIP) in SpA.46,47 These data as well as the diagnostic 2 value of this autoantibody should be validated in independent cohorts.

In summary, despite major efforts over the last years only few biomarkers have been described and validated in SpA, and often these biomarkers are useful at the group level but lack specificity to be used in individual patients in clinical care. Interestingly, SpAis genetically, pathophysiologically and clinically associated with psoriasis and inflammatory bowel diseases (IBD), two conditions in which serum biomarkers have been described. Therefore, we reviewed here the value of biomarkers in these conditions and discuss their potential utility for SpA. Reviewing was done for the settings of diagnosis (i.e. difference between disease and healthy or disease and similar inflammatory diseases), with respect to correlation with disease activity and response to effective treatment, as well asfor predicting structural damage (in axial SpA). We have evaluated both the pathophysiological and clinical value of these markers.

SERUM BIOMARKERS IN PSORIASIS

Psoriasis is an immune-mediated, chronic inflammatory disease primarily of the skin but can also affect nails and peripheral and axial joints. The disease is characterized by abnormal differentiation and hyperproliferation of epidermal keratinocytes and infiltration of immune cells. Roughly 1 in 5 psoriasis patients will develop psoriatic arthritis (PsA), one ofthe phenotypical forms of SpA.48 At the genetic level, SNPs in IL-23R and Th-17 genes have shown to be associated with both psoriasis and SpA.

In clinical practice, a diagnosis of psoriasis is based on clinical history and examination and, in case of diagnostic uncertainty, skin biopsies. Disease activity is mainly determined by the Psoriasis Area and Severity Index (PASI),49 which includes the percentage of affected skin area, the erythema, the induration, and scaling. Currently, serum biomarkers do not play an important role in clinical practice but are extensively studied in clinical trials (Table 1).

Serum CRP levels were found to be elevated in the moderate and severe forms of psoriasis but not in patients with mild disease, when compared with the healthy individuals.50,51 Several studies have reported CRP elevation in 17-45.7% of the psoriasis patients versus 1% in the control group.51–53 Furthermore, CRP levels decreased significantly after effective treatment was started.51,54–57 It is therefore only a potentially useful marker for disease activity in patients who did not receive systemic treatment for at least one month. In the treated patients, PASI is preferred to monitor disease activity. ESR levels were elevated in

CHAPTER 2 23 Table 1. Serum biomarkers in psoriasis: elevation in comparison with healthy controls, correlation with disease activity according to PASI, and modulation by effective treatment Elevation in Correlation with comparison with disease activity Biomarker healthy controls measured PASI Treatment response References CRP + +/- +/- [51–55] ESR + NA NA [58] TNFα + NA + [59, 60] TNFα-R1 + +/- NA [59, 66, 77, 99] IFNγ + NA NA [60] IL-8 + - NA [59, 60, 63] IL-12 +/- + NA [60] IL-17 +/- + + [59, 60, 64, 65] IL-18 + + + [60, 61, 70] IL-22 + + + [59, 62, 65] E-selectin + +/- NA [63, 66–68] ICAM-1 + + +/- [63, 66, 67, 69, 70] VAP-1 + - - [72–74] S100A8/A9 (calprotectin) + + NA [75, 76] LCN-2 + +/- - [77–80] hBD-2 + + +/- [81, 82] Resistin + +/- +/- [83, 90, 91] Leptin + - - [83–87, 90] Adiponectin - +/- +/- [84–86, 88, 90, 92] Ghrelin + + - [89, 90] RBP-4 + + - [80, 85] VEGF + + +/- [59, 62, 93–95] Nitric oxide + +/- + [94, 96–98] CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; hBD-2, human beta-defensin-2; ICAM-1, intracellular adhesion molecule-1; IFNγ, interferon gamma; IL, interleukin; LCN-2, lipocalin-2; NA, not applicable; PASI, Psoriasis Area and Severity Index; RBP-4, retinol binding protein-4; TNF, tumor necrosis factor; TNFα-R1, tumor necrosis factor alpha-receptor 1; VEGF, vascular endothelial growth factor; VAP-1, vascular adhesion protein-1.

approximately half of the psoriasis patients compared to healthy controls. Elevated ESR was clearly more frequent in the active state of the disease (80%) versus non-active psoriasis.58

Cytokines, including interferon-gamma (IFNγ), TNF-α, TNF-R1, IL-6, IL-8, IL-12, and IL-18, IL- 22, and IL-23 have been investigated thoroughly in psoriasis. Coimbra et al. and others59–63 have found that the serum levels of these cytokines were overall significantly higher in psoriasis as compared with healthy controls. But the data on the correlation with PASI were rather conflicting. Psoriasis-specific treatment including methotrexate, UV-A/UV-B treatment, and TNF blockade did have an effect on serum levels of these cytokines.59,60,64,65

Adhesion molecules are expressed by activated tissue cells (e.g. endothelium) as well as leukocytes and allow selective migration of distinct leukocyte subsets to specific tissues and

24 SERUM BIOMARKERS IN PSORIASIS AND IBD sites of inflammation. For example, activated epithelial cells express E-selectin, which can interact with E-selectin on another epithelial cell or with alpha E beta 7 (a member of the integrin family) on CD8 positive T cells. Of the soluble adhesion molecules that can easily be measured in serum, E-selectin, ICAM-1, and VAP-1 are best studied in psoriasis.63,66–74 All of 2 these molecules showed elevated serum levels in psoriasis versus healthy controls. Soluble E-selectin and ICAM-1 levels, however, were consistently (but only weakly to moderately) associated with disease activity as expressed by PASI scores (r=0.44 and r=0.26-0.43, respectively). Accordingly, only E-selectin levels decreased significantly and consistently upon effective treatment.63,67,68,71,72

Among the antimicrobial peptides and proteins (AMP), calprotectin (S100A8/A9 proteins), lipocalin-2 (LCN-2, also called neutrophil gelatinase-associated lipocalin [NGAL]), and human β-defensin-2 (hBD-2) were associated with psoriasis. Highly up-regulated calprotectin levels were found in a proteomic analysis of the epidermis from psoriasis patients.75 In addition, Benoit et al.76 showed that calprotectin levels in serum were significantly elevated when compared to healthy controls and that patients with a high PASI (>15) score also had higher levels of calprotectin. Regarding LCN-2 serum levels, several groups consistently found elevated levels in psoriasis patients when compared with healthy individuals. However, absent or only weak (r=0.27) correlations were found between LCN-2 levels and PASI scores, and serum levels were not normalized by effective treatment.77–80 Serum levels of another AMP, hBD-2, were elevated consistently when compared with healthy donors. Jansen et al.81 reported a high correlation with PASI (r=0.82), whereas another group did not see such a strong correlation (r<0.1, p>0.05).82

Adipokines are produced by fat (adipose) tissue and play a role in metabolic processes, but these proteins are also associated with inflammation. Extensively studied adipokines were resistin, leptin, adiponectin, ghrelin, and retinol binding protein-4 (RBP-4). With the exception of adiponectin, all serum and plasma adipokine levels were significantly higher in psoriasis versus healthy individuals.80,83–89 Only resistin levels correlated weakly with PASI (r2=0.027) and none of the adipokines consistently showed a normalization in levels after effective treatment.80,85,88,90–92

Growth factors are involved in angiogenesis and hypervascularity in psoriasis and are well studied as biomarkers. Of these growth factors, VEGF was the most evaluated markers. VEGF was the most promising in psoriasis, since there were differences in serum levels between psoriasis patients and healthy controls and there was a strong correlation (R=0.65) with PASI.59,93,94 However, serum levels did not normalize upon effective treatment.59,62,95

Nitric oxide (NO) is a free radical and is secreted by various cell types, including fibroblasts and endothelial cells. Tekin et al.96 performed a study in 22 psoriasis samples versus 21 healthy controls and found significantly higher serum levels in the former group. These results were reproduced by others.94,97,98 Furthermore, NO plasma and serum levels

CHAPTER 2 25 decreased after effective treatment.96

In summary, CRP, ESR, IL-18, IL-22, ICAM-1, calprotectin, LCN-2, hBD-2, VEGF, RBP-4, and NO were the most informative biomarkers since these markers are elevated in patients with the disease when compared to healthy controls, show weak to moderate correlations with disease activity, and/or decrease upon effective treatment at the group level, indicating that -at least from a pathophysiological point of view- these are interesting markers. In spite of this, though, the value of these markers as biomarker in clinical practice (at the individual patient level) seems –if investigated- modest at best. In psoriasis, CRP and ESR were the only serum biomarkers used in clinical practice for the evaluation of treatment, despite the disappointing results earlier mentioned. The most interesting serum and plasma biomarkers in psoriasis from a pathophysiological point of view are the antimicrobial peptides hBD-2 and LCN-2: the expression of both peptides is driven by IL-17 and may therefore reflect a key pathophysiological process in psoriasis skin, although the role in clinical practice remains inconclusive.

SERUM BIOMARKERS IN IBD

IBD includes Crohn’s disease (CD) and ulcerative colitis (UC). Peripheral arthritis and sacroiliitis are seen in 7-16%100–102 and 2-32%100,102,103 of the IBD patients, respectively. Similarly, 60- 70% of the SpA patients show subclinical signs of gut inflammation on biopsies obtained by ileocolonscopy,104,105 but only 7% of these patients will eventually develop clinically overt IBD over time. The link between SpA and IBD is corroborated by genetic studies, with the best example being polymorphisms in the interleukin 23 receptor (IL23R) gene which are associated both with IBD (OR 0.38 and 0.73 for CD and UC, respectively) and AS (OR=0.53- 1.27).106–109

The diagnosis of IBD in clinical practice is based on the combination of clinical history, physical examination, stool findings (i.e. calprotectin and lactoferrin), serological antibodies, inflammatory markers in blood, and colonoscopy.110 The major clinical outcome measure in Crohn’s disease, the Crohn’s Disease Activity Index (CDAI),111 includes clinical and laboratory findings to determine disease activity. In UC, several scores are used to determine disease activity including clinical and endoscopic scores. A frequently used clinical score is Simple Clinical Colitis Activity Index (SCCAI).112 The Mayo activity index is based on clinical as well as endoscopic findings.113

Of the acute phase reactants, CRP was found to be a biomarker for IBD since the serum levels were elevated in CD and to a lesser extent in UC when compared with healthy controls or functional bowel disorders.114,115 Elevated levels were correlated with a higher disease activity.116–118 Louis et al.118 studied a group of 226 CD patients receiving TNF-inhibitors and

26 SERUM BIOMARKERS IN PSORIASIS AND IBD a clinical response was associated with a decrease in CRP with similar results in a larger cohort. There were discrepant data about the ability of CRP to predict disease relapse.119,120 Other well studied inflammatory markers were ESR, procalcitonin, and YKL-40. All three markers correlated with the disease activity but only ESR and procalcitonin differentiated 2 between IBD and healthy individuals or other gastrointestinal diseases.121–128 Moreover, a trend towards elevated levels of YKL-40 was found in more severe CD patients having strictures (i.e. narrowing of the intestinal lumina),124 but since the majority of active patients do not have elevated levels, in clinical practice YKL-40 is of little value.

Cytokines measured in serum did not seem to be important markers of disease in IBD, despite their role in the inflammatory process.129,130 The only potential marker is IL-6, the main driver of CRP. Levels of IL-6 in serum were increased in active CD compared with healthy controls but correlation with disease activity was weak.131–133

Of the soluble adhesion molecules, ICAM-1 was the most promising serum and plasma marker as the levels were elevated when compared with healthy controls. Although discrepant data are available on correlation with disease activity, effective treatment down- regulates ICAM-1 levels.134–138 E-selectin levels in serum or plasma were significantly higher in IBD than in healthy controls. However, there was no significant association between disease activity and treatment.134–138 A few studies were published on vascular cell adhesion molecule (VCAM). The serum or plasma levels in IBD were not consistently elevated versus healthy controls and correlations to disease activity were poor.134–136

The adipokines adiponectin, leptin, ghrelin, resistin, and vaspin had been investigated thoroughly in IBD. Only serum levels of resistin and vaspin were consistently elevated in IBD versus healthy individuals. Moreover, strong correlations with PASI were found for ghrelin, resistin and vaspin levels with disease activity.139–147 Ghrelin and resistin levels were decreased upon effective treatment but this was not studied for the other adipokines.148,149

Of the S100 proteins, S100A12 levels in serum were elevated in IBD when compared with healthy controls, but there are conflicting data about the correlation with active disease activity.150,151 Serum levels of S100A8/S100A9, calprotectin, were elevated in IBD versus healthy individuals and correlated with the CDAI.152–154 Similar results were seen in children using pediatric CDAI indices.155 Furthermore, calprotectin serum levels decreased upon treatment with TNF inhibition.154,156 Fecal calprotectin is one of the most extensively studied stool markers in IBD. Calprotectin levels in stool can differentiate between IBD and healthy controls or irritable bowel syndrome (IBS). One meta-analysis of Von Roon et al.157 included 30 studies with 1210 IBD patients, 297 colorectal cancer, 697 IBS, and 3393 healthy controls and reported that fecal calprotectin is a useful biomarker to differentiate between IBD and non-IBD patients. The overall weighted area under curve (AUC) was with 0.95-0.98 (using a cut-off level of 50 ug/g and 100ug/g, respectively) exceptionally high. Moreover, differentiating CD from IBS by fecal calprotectin showed an AUC of 0.97. Van Rheenen et

CHAPTER 2 27 al.158 determined that fecal calprotectin could be used as screening tool in suspected clinical IBD, thereby decreasing the use of invasive colonoscopy by 67%, but at the cost of false negative results in 6% of the cases. Results of a meta-analysis that included 13 studies (1471 IBD patients), showed that fecal calprotectin is also a good marker for disease activity based on the Mayo activity index.159 Moreover, fecal calprotectin decreased upon treatment160–163 and may predict disease relapse.120,164,165

Several AMPs have been investigated in IBD. Serum LCN-2 (NGAL) levels were elevated in IBD when compared with healthy controls and IBS.166–168 HBD-2 seemed a potential marker since expression levels were increased in inflamed mucosa versus non-inflamed mucosa of the colon.169 However, Yamaguchi et al. found no differences in hBD-2 plasma levels in UC or CD when compared to healthy controls.170

Beta-2-microglobulin (β2-M), which forms a part of the major histocompatibility complex class 1 (MHC 1), had been investigated in two independent studies. In 2001 Zissis et al.171 showed that serum levels were elevated in CD versus healthy controls, and that serum levels were higher when patients had a higher disease activity, but they did not see similar results for UC. Yilmaz et al.172 recently showed similar results but now in the whole IBD population.

Of the serological antibodies studied in IBD,173–177 anti-saccharomyces cervisiae antibodies (ASCA) and perinuclear anti-neutrophil cytoplasmic antibodies (pANCA) showed the highest diagnostic accuracy. Sensitivity and specificity for ASCA were 60% and 91% and for pANCA 50% and 95% respectively, when distinguishing IBD from healthy controls. Sensitivity and specificity for distinguishing between CD and UC were for ASCA (CD): 60% and 86%, and for pANCA (UC): 50% and 94% respectively.178 Similar results were seen in other studies.179,180 Also, when discriminating CD from UC or when comparing with other gastrointestinal diseases, including celiac disease, the specificity was too low for these antibodies.174–178 Both autoantibodies had a positive predictive value of 80-90% and a negative predictive value of 50-80%. The number of false-positive and false-negative individual was too high to recommend these autoantibodies for routine diagnostic procedures. Ester et al.181 found no association of ASCA or pANCA with response to infliximab treatment, but ASCA positivity was associated with worse disease progression: a more severe disease type (from inflammation to structuring (or stenosis) and penetrating) disease in CD.182 Recently, Bonneau et al.183 published a systematic review in which anti-glycan, anti-GP2, and anti-GM-CSF antibodies were also described as having low sensitivity but high specificity to CD when compared with UC but the contribution of these antibodies to diagnosis is not yet clear.

To summarize, CRP and ESR, the S100 family proteins S100A12 and S100A8/A9 (calprotectin), the antimicrobial peptides LCN-2 and hBD-2, the MHC class 1 molecule β2-M, and ASCA antibodies are of interest as biomarkers in IBD from a pathophysiological point ofview. In clinical practice however, only a few of these markers are used, including CRP, ESR, and fecal calprotectin. Since CRP correlates with disease activity and CRP levels normalize upon

28 SERUM BIOMARKERS IN PSORIASIS AND IBD effective treatment, evaluation of therapy by CRP is often applied in general practice. Fecal calprotectin, but not serum calprotectin, is extensively used in clinical practice to identify IBD patients with intestinal inflammation, providing high sensitivity and specificity in detecting IBD. Nevertheless, the ileocolonoscopy remains the gold standard. With regard 2 to antibodies, ASCA antibodies are able to differentiate between CD and UC and are used in clinical practice although not as a screening diagnostic procedure. Other markers in IBD that are of pathophysiological interest are the IL-17 driven LCN-2, hBD-2, and β2-M, that forms part of the MHC class-I molecule (Table 2).

Table 2. Serum biomarkers in IBD levels: elevation in comparison with healthy controls, correlation with disease activity according to CDAI, SCCAI, or endoscopic activity, modulation by effective treatment, and outcome of structural damage Correlation with Elevation In disease activity comparison measured by CDAI or Outcome with healthy SCCAI or endoscopic Treatment structural Biomarker controls activity response damage References CRP + + + + [114–121] ESR + + NA NA [114, 128] Procalcitonin + +/- NA NA [125–127] YKL-40 +/- + NA +/- [122–124] IL-6 + +/- NA NA [131–133, 156] ICAM-1 +/- - + NA [134–138] E-selectin + - - NA [134–138] VCAM +/- - - NA [134–136] Adiponectin + - - NA [145–147, 149, 184] Leptin +/- - +/- NA [145, 146, 148, 149, 184] Ghrelin + + + +/- [140, 141, 146, 184, 185] Resistin + +/- + NA [142, 143, 146, 149, 184] Vaspin +/- - NA NA [144] S100A12 + +/- + NA [150, 151] S100A8/A9 + NA NA + [152–155] LCN-2 + + NA NA [166–168] hBD-2 - - NA NA [169, 170] β2-M + + NA NA [171, 172] [173, 175, 177, 178, 181, ASCA + + - + 182] pANCA + - + - [173–178, 181, 182] ASCA, anti-saccharomyces cerevisiae antibodies; β2-M, β2-microglobulin; CDAI, crohn’s disease activity index; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; hBD-2, human beta-defensin-2; ICAM, intracellular adhesion molecule-1; IL, interleukin; LCN-2, lipocalin-2; NA, not applicable; pANCA, perinuclear anti-neutrophil cytoplasmic antibodies; SCCAI, simple clinical colitis activity index; VCAM, vascular cell adhesion molecule; VEGF, vascular endothelial growth factor.

CHAPTER 2 29 LESSONS TO BE LEARNED FROM POTENTIAL SERUM BIOMARKERS IN PSORIASIS AND IBD

Several biomarkers studied in psoriasis and IBD may be informative in SpA as well. CRP, ESR, calprotectin, and VEGF are interesting in psoriasis and IBD. These markers have already been studied in SpA (as mentioned above). While they may be discriminatory at the group level, sensitivity and/or specificity are too low to be used in clinical practice for diagnosis, disease activity measurement, evaluation of treatment response, or prediction of axial spinal progression.

Some of the biomarkers of interest in in psoriasis and IBD have also been studied in SpA. Negative or variable results were found for serum-IL-18,186 ICAM-1,187–189 RBP-4,190,191 and NO.192,193

ASCA antibodies can help to discriminate CD from UC, have already been studied in SpA. The levels of ASCA antibodies were found to be higher in SpA patients than in healthy controls, as reported by different studies.194,195 One study determined these antibodies in non-IBD SpA patients, and 30% of them appeared to be ASCA-positive. Nevertheless, levels did not correlate with disease activity.196

The AMPs are a new family of relevant biomarkers in psoriasis and IBD as was found in several studies. AMPs have not yet been extensively studied in SpA. The most interesting members of this family are LCN-2 and hBD-2, both of which are driven by IL-17 and are reflecting active disease in psoriasis as well as IBD. And because the reliable measurement of IL-17 in serum remains a challenge, these biomarkers may be used as a proxy for IL-17-mediated tissue inflammation in psoriasis and IBD. As SpA is also an IL-17-driven disease,197–201 these markers certainly deserve further study in this condition. It should be noted, however, that these AMPs are mainly produced by epithelial cells and may thus be more useful in skin and gut diseases than in other IL-17-related inflammatory conditions. Along the same line, IL-22 is an IL-10 cytokine family member which is produced by Th17 cells under the control of IL-23. It was shown that plasma IL-22 levels were increased in SpA versus healthy individuals, but no correlation was found with Bath ankylosing spondylitis disease activity index (BASDAI).202 The reliability of serum IL-22 measurement and the validity of these findings require replication in independent studies.

Another biomarker which was identified in IBD and may be directly related to the pathological processes in SpA is β2-M. β2-M is a component of class I MHC molecules, including HLA-B27, which is (still) the major genetic risk factor for SpA. Although the exact role of HLA-B27 in the disease pathogenesis is not yet clear, several hypotheses suggest that the level of expression of β2-M may modulate the pathological role of HLA-B27. One hypothesis proposes that β2-

30 SERUM BIOMARKERS IN PSORIASIS AND IBD M-free HLA-B27 heavy chains form disulphide-linked homodimers which can be recognized by the KIR3DL2 receptors on NK-cells and T-cells.203 The second hypothesis concerns HLA-B27 heavy chain misfolding prior to binding of β2-M and peptide, resulting in unfolded protein responses (UPR), which subsequently may lead to an inflammatory cytokine release.201,204 2 Interestingly, modulation of β2-M expression levels in the HLA-B27 transgenic rat model of SpA did profoundly affect the phenotype of the disease,205–207 confirming the potential importance of this molecule in SpA. Therefore, serum β2-M expression levels should be further explored as potential biomarker in SpA.

EXPERT COMMENTARY

To conclude, of the serum biomarkers identified in psoriasis and IBD, the IL-17 driven anti- microbial peptides hBD-2 and LCN-2, and the class I MHC molecule β2-microglobulin are the most interesting candidates from a pathophysiological standpoint to be further explored in SpA for clinical purposes.

KEY POINTS

• SpA is an immune-mediated inflammatory disease affecting the spine, the peripheral joints, and extra-articular tissues such as the skin, the gut, and the eye. • Serum biomarkers could address several challenges in SpA (diagnosis, disease activity monitoring, treatment response, and outcome). • Serum biomarkers well studied in spondyloarthritis are CRP, ESR, IL-6, calprotectin, VEGF, MMP-3, DKK-1, and sclerostin. • Although the aforementioned biomarkers are useful at the group level in SpA, these markers lack specificity for the use in individual patients in clinical care. • SpA is a clinically and pathophysiologically closely related to psoriasis and inflammatory bowel disease. • In psoriasis, CRP, ESR, IL-18, IL-22, ICAM-1, calprotectin, LCN-2, hBD-2, VEGF, RBP-4, and NO were the most interesting markers from a pathophysiological standpoint. • In IBD, CRP and ESR, hBD-2, LCN-2, and ASCA antibodies were of interest from a pathophysiological point of view. • HBD-2, LCN-2, and β2-microglobulin are the most interesting candidates from a pathophysiological standpoint to be further explored in SpA for clinical purposes.

CHAPTER 2 31 REFERENCES

1. Dougados M, Baeten D: Spondyloarthritis. Lancet 2014, 73:e45–e45. 2011, 377:2127–37. 12. Gruttola VG De, Clax P, Demets DL, Downing GJ, 2. Sieper J, Rudwaleit M, Baraliakos X, Brandt J, Braun Ellenberg SS, Friedman L, Gail MH, Prentice R, Wittes J, J, Burgos-Vargas R, Dougados M, Hermann K-G, Zeger SL: Considerations in the Evaluation of Surrogate Landewé R, Maksymowych W, van der Heijde D: The Endpoints in Clinical Trials: Summary of a National Assessment of SpondyloArthritis international Society Institutes of Health Workshop.Control Clin Trials 2001, (ASAS) handbook: a guide to assess spondyloarthritis. 1998:485–502. Ann Rheum Dis 2009, 68 Suppl 2:ii1–44. 13. Rudwaleit M, van der Heijde D, Landewé R, Akkoc N, 3. Van Den Bosch F, Kruithof E, Baeten D, Herssens A, de Brandt J, Chou CT, Dougados M, Huang F, Gu J, Kirazli Keyser F, Mielants H, Veys EM: Randomized double- Y, Van den Bosch F, Olivieri I, Roussou E, Scarpato S, blind comparison of chimeric monoclonal antibody Sørensen IJ, Valle-Oñate R, Weber U, Wei J, Sieper J: to tumor necrosis factor alpha (infliximab) versus The Assessment of SpondyloArthritis International placebo in active spondylarthropathy. Arthritis Rheum Society classification criteria for peripheral 2002, 46:755–65. spondyloarthritis and for spondyloarthritis in general. 4. Braun J, Brandt J, Listing J, Zink a, Alten R, Golder W, Ann Rheum Dis 2011, 70:25–31. Gromnica-lhle E, Kellner H, Krause a, Schneider M, 14. Benhamou M, Gossec L, Dougados M: Clinical Sörensen H, Zeidler H, Thriene W, Sieper J: Treatment relevance of C-reactive protein in ankylosing of active ankylosing spondylitis with infliximab: a spondylitis and evaluation of the NSAIDs/coxibs’ randomised controlled multicentre trial. Lancet 2002, treatment effect on C-reactive protein. Rheumatology 359:1187–1193. (Oxford) 2010, 49:536–41. 5. Mease PJ, Goffe BS, Metz J, VanderStoep A, Finck B, 15. Poddubnyy D a, Rudwaleit M, Listing J, Braun J, Sieper Burge DJ: Etanercept in the treatment of psoriatic J: Comparison of a high sensitivity and standard arthritis and psoriasis: a randomised trial. Lancet C reactive protein measurement in patients with 2000, 356:385–390. ankylosing spondylitis and non-radiographic axial 6. Paramarta JE, De Rycke L, Heijda TF, Ambarus C a, Vos spondyloarthritis. Ann Rheum Dis 2010, 69:1338–41. K, Dinant HJ, Tak PP, Baeten DL: Efficacy and safety of 16. Visvanathan S, Wagner C, Marini JC, Baker D, Gathany adalimumab for the treatment of peripheral arthritis T, Han J, van der Heijde D, Braun J: Inflammatory in spondyloarthritis patients without ankylosing biomarkers, disease activity and spinal disease spondylitis or psoriatic arthritis. Ann Rheum Dis 2013, measures in patients with ankylosing spondylitis 72:1793–9. after treatment with infliximab. Ann Rheum Dis 2008, 7. Sieper J, van der Heijde D, Dougados M, Mease PJ, 67:511–7. Maksymowych WP, Brown M a, Arora V, Pangan AL: 17. De Vries MK, van Eijk IC, van der Horst-Bruinsma Efficacy and safety of adalimumab in patients with IE, Peters MJL, Nurmohamed MT, Dijkmans B non-radiographic axial spondyloarthritis: results of a a C, Hazenberg BPC, Wolbink GJ: Erythrocyte randomised placebo-controlled trial (ABILITY-1). Ann sedimentation rate, C-reactive protein level, and Rheum Dis 2013, 72:815–22. serum amyloid a protein for patient selection and 8. Landewé R, Braun J, Deodhar a, Dougados M, monitoring of anti-tumor necrosis factor treatment Maksymowych WP, Mease PJ, Reveille JD, Rudwaleit in ankylosing spondylitis. Arthritis Rheum 2009, M, van der Heijde D, Stach C, Hoepken B, Fichtner 61:1484–90. a, Coteur G, de Longueville M, Sieper J: Efficacy of 18. Poddubnyy D, Rudwaleit M, Haibel H, Listing J, Märker- certolizumab pegol on signs and symptoms of axial Hermann E, Zeidler H, Braun J, Sieper J: Rates and spondyloarthritis including ankylosing spondylitis: 24- predictors of radiographic sacroiliitis progression over week results of a double-blind randomised placebo- 2 years in patients with axial spondyloarthritis. Ann controlled Phase 3 study. Ann Rheum Dis 2014, 73:39– Rheum Dis 2011, 70:1369–74. 47. 19. Gratacós J, Collado A, Filella F, Sanmartí R, Cañete J, 9. Feldtkeller E, Khan MA, van der Heijde D, van der Llena J, Molina R, Ballesta A, Muñoz-Gómez J: Serum Linden S, Braun J: Age at disease onset and diagnosis cytokines (IL-6, TNF-alpha, IL-1 beta and IFN-gamma) delay in HLA-B27 negative vs. positive patients with in ankylosing spondylitis: a close correlation between ankylosing spondylitis. Rheumatol Int 2003, 23:61–6. serum IL-6 and disease activity and severity. Br J 10. Salvadorini, F. Bandinelli, A. Delle Sedie, L. Riente, A. Rheumatol 1994, 33:927–931. Candelieri, S. Generini, N. Possemato, S. Bombardieri 20. Pedersen SJ, Hetland ML, Sørensen IJ, Ostergaard MM-C: Ankylosing spondylitis: how diagnostic and M, Nielsen HJ, Johansen JS: Circulating levels of therapeutic delay have changed over the last six interleukin-6, vascular endothelial growth factor, YKL- decades. Clin Exp Rheumatol 2012, 30:561–565. 40, matrix metalloproteinase-3, and total aggrecan in 11. Sorensen J, Hetland ML: Decreases in diagnostic delay spondyloarthritis patients during 3 years of treatment are supported by sensitivity analyses. Ann Rheum Dis with TNFα inhibitors. Clin Rheumatol 2010, 29:1301–

32 SERUM BIOMARKERS IN PSORIASIS AND IBD 9. matrix metalloproteinases and tissue inhibitors of 21. Kane D, Roth J, Frosch M, Vogl T, Bresnihan B, metalloproteinases in ankylosing spondylitis: MMP-3 FitzGerald O: Increased perivascular synovial is a reproducibly sensitive and specific biomarker of membrane expression of myeloid-related proteins in disease activity.Rheumatology (Oxford) 2006, 45:414– psoriatic arthritis. Arthritis Rheum 2003, 48:1676–85. 20. 22. Kruithof E, De Rycke L, Vandooren B, De Keyser F, 32. Yang C, Gu J, Rihl M, Baeten D, Huang F, Zhao M, 2 FitzGerald O, McInnes I, Tak PP, Bresnihan B, Veys Zhang H, Maksymowych WP, De Keyser F, Veys EM, EM, Baeten D: Identification of synovial biomarkers Yu DTY: Serum levels of matrix metalloproteinase 3 of response to experimental treatment in early-phase and macrophage colony-stimulating factor 1 correlate clinical trials in spondylarthritis.Arthritis Rheum 2006, with disease activity in ankylosing spondylitis.Arthritis 54:1795–804. Rheum 2004, 51:691–9. 23. De Rycke L, Baeten D, Foell D, Kruithof E, Veys EM, Roth 33. Pedersen SJ, Sørensen IJ, Lambert RGW, Hermann J, De Keyser F: Differential expression and response to K-G a, Garnero P, Johansen JS, Madsen OR, Hansen anti-TNFalpha treatment of infiltrating versus resident A, Hansen MS, Thamsborg G, Andersen LS, Majgaard tissue macrophage subsets in autoimmune arthritis. J O, Loft AG, Erlendsson J, Asmussen KH, Jurik AG, Pathol 2005, 206:17–27. Møller J, Hasselquist M, Mikkelsen D, Østergaard M: Radiographic progression is associated with resolution 24. Turina MC, Sieper J, Yeremenko N, Conrad K, Haibel of systemic inflammation in patients with axial H, Rudwaleit M, Baeten D, Poddubnyy D: Calprotectin spondylarthritis treated with tumor necrosis factor α serum level is an independent marker for radiographic inhibitors. Arthritis Rheum 2011, 63:3789–800. spinal progression in axial spondyloarthritis. Ann Rheum Dis 2014, 73:1746–8. 34. Maksymowych WP, Landewé R, Conner-Spady B, Dougados M, Mielants H, van der Tempel H, 25. Baeten DL, Demetter P, Cuvelier C, Bosch F Van Poole a R, Wang N, van der Heijde D: Serum matrix den, Kruithof E, Damme N van, Verbruggen G, metalloproteinase 3 is an independent predictor Mielants H, Veys E, Keyser F De: Comparative study of structural damage progression in patients with of the synovial histology in rheumatoid arthritis , ankylosing spondylitis. Arthritis Rheum 2007, spondyloarthropathy , and osteoarthritis : influence 56:1846–53. of disease duration and activity. Ann Rheum Dis 2000, 59:945–953. 35. Diarra D, Stolina M, Polzer K, Zwerina J, Ominsky MS, Dwyer D, Korb A, Smolen J, Hoffmann M, Scheinecker 26. Baeten D, Kruithof E, De Rycke L, Vandooren B, Wyns B, C, van der Heide D, Landewe R, Lacey D, Richards WG, Boullart L, Hoffman IE a, Boots AM, Veys EM, De Keyser Schett G: Dickkopf-1 is a master regulator of joint F: Diagnostic classification of spondylarthropathy remodeling. Nat Med 2007, 13:156–63. and rheumatoid arthritis by synovial histopathology: a prospective study in 154 consecutive patients. 36. Klingberg E, Nurkkala M, Carlsten H, Forsblad-d’Elia Arthritis Rheum 2004, 50:2931–41. H: Biomarkers of bone metabolism in ankylosing spondylitis in relation to osteoproliferation and 27. Drouart M, Saas P, Billot M, Cedoz J-P, Tiberghien osteoporosis. J Rheumatol 2014, 41:1349–56. P, Wendling D, Toussirot E: High serum vascular endothelial growth factor correlates with disease 37. Kwon S-R, Lim M-J, Suh C-H, Park S-G, Hong Y-S, Yoon activity of spondylarthropathies. Clin Exp Immunol B-Y, Kim H-A, Choi H-J, Park W: Dickkopf-1 level is lower 2003, 132:158–62. in patients with ankylosing spondylitis than in healthy people and is not influenced by anti-tumor necrosis 28. Poddubnyy D, Conrad K, Haibel H, Syrbe U, Appel H, factor therapy. Rheumatol Int 2012, 32:2523–7. Braun J, Rudwaleit M, Sieper J: Elevated serum level of the vascular endothelial growth factor predicts 38. Heiland GR, Appel H, Poddubnyy D, Zwerina J, Hueber radiographic spinal progression in patients with axial A, Haibel H, Baraliakos X, Listing J, Rudwaleit M, Schett spondyloarthritis. Ann Rheum Dis 2014, 73:2137–43. G, Sieper J: High level of functional dickkopf-1 predicts protection from syndesmophyte formation in patients 29. Baraliakos X, Hermann KG, Xu S, Hsu B, Braun J: VEGF with ankylosing spondylitis. Ann Rheum Dis 2012, and CRP serum levels lack predictive value for outcomes 71:572–4. as assessed by conventional radiographs and magnetic resonance imaging in patients with active ankylosing 39. Appel H, Ruiz-Heiland G, Listing J, Zwerina J, Herrmann spondylitis treated with the TNF inhibitor golimumab. M, Mueller R, Haibel H, Baraliakos X, Hempfing A, Ann Rheum Dis 2014;73(Suppl2):1138-1139. Rudwaleit M, Sieper J, Schett G: Altered skeletal expression of sclerostin and its link to radiographic 30. Vandooren B, Kruithof E, Yu DTY, Rihl M, Gu J, De progression in ankylosing spondylitis. Arthritis Rheum Rycke L, Van Den Bosch F, Veys EM, De Keyser F, 2009, 60:3257–62. Baeten D: Involvement of matrix metalloproteinases and their inhibitors in peripheral synovitis and down- 40. Taylan A, Sari I, Akinci B, Bilge S, Kozaci D, Akar S, regulation by tumor necrosis factor alpha blockade in Colak A, Yalcin H, Gunay N, Akkoc N: Biomarkers and spondylarthropathy. Arthritis Rheum 2004, 50:2942– cytokines of bone turnover: extensive evaluation in a 53. cohort of patients with ankylosing spondylitis. BMC Musculoskelet Disord 2012, 13:191. 31. Chen C-H, Lin K-C, Yu DTY, Yang C, Huang F, Chen H, Liang T-H, Liao H-T, Tsai C-Y, Wei JCC, Chou C-T: Serum 41. De Andrade KR, de Castro GRW, Vicente G, da Rosa JS, Nader M, Pereira IA, Fröde TS: Evaluation of circulating

CHAPTER 2 33 levels of inflammatory and bone formation markers in Dermatol 2008, 159:322–30. axial spondyloarthritis. Int Immunopharmacol 2014, 55. Serwin AB, D M, Wasowicz W, Ph D, Chodynicka B, 21:481–6. D M: Selenium supplementation , soluble tumor 42. Korkosz M, Jerzy G, Leszczy P, Pawlak-bu K, Jeka necrosis factor-alpha receptor type 1 , and C-reactive S, Kucharska E: High disease activity in ankylosing protein during psoriasis therapy with narrowband spondylitis is associated with increased serum ultraviolet B. Nutrition 2006, 22:860–864. sclerostin level and decreased wingless protein-3a 56. Coimbra S, Oliveira H, Reis F, Belo L, Rocha S, signaling but is not linked with greater structural Quintanilha a., Figueiredo a., Teixeira F, Castro E, damage. BMC Musculoskelet Disord 2013;14:99. Rocha-Pereira P, Santos-Silva a.: Circulating adipokine doi:10.1186/1471-2474-14-99. levels in Portuguese patients with psoriasis vulgaris 43. Chou C-L, Wu M-J, Yu C-L, Lu M-C, Hsieh S-C, Wu according to body mass index, severity and therapy. J T-H, Chou C-T, Tsai C-Y: Anti-agalactosyl IgG antibody Eur Acad Dermatology Venereol 2010, 24:1386–1394. in ankylosing spondylitis and psoriatic arthritis. Clin 57. Gisondi P, Lora V, Bonauguri C, Russo a, Lippi G, Rheumatol 2010, 29:875–81. Girolomoni G: Serum chemerin is increased in patients 44. Wright C, Sibani S, Trudgian D, Fischer R, Kessler with chronic plaque psoriasis and normalizes following B, Labaer J, Bowness P: Detection of Multiple treatment with infliximab. Br J Dermatol 2013, Autoantibodies in Patients with Ankylosing Spondylitis 168:749–55. Using Nucleic Acid Programmable Protein Arrays. Mol 58. Rocha-Pereira P, Santos-Silva a, Rebelo I, Figueiredo a, Cell Proteomics 2012:1–10. Quintanilha a, Teixeira F: The inflammatory response 45. Duftner C, Dejaco C, Klauser a, Falkenbach a, in mild and in severe psoriasis. Br J Dermatol 2004, Lakomek HJ, Schirmer M: High positive predictive 150:917–28. value of specific antibodies cross-reacting with a 28- 59. Coimbra S, Oliveira H, Reis F, Belo L, Rocha S, kDa Drosophila antigen for diagnosis of ankylosing Quintanilha a, Figueiredo a, Teixeira F, Castro E, spondylitis. Rheumatology (Oxford) 2006, 45:38–42. Rocha-Pereira P, Santos-Silva a: Interleukin (IL)-22, 46. Baerlecken NT, Nothdorft S, Stummvoll GH, Sieper J, IL-17, IL-23, IL-8, vascular endothelial growth factor Rudwaleit M, Reuter S, Matthias T, Schmidt RE, Witte and tumour necrosis factor-α levels in patients with T: Autoantibodies against CD74 in spondyloarthritis. psoriasis before, during and after psoralen-ultraviolet Ann Rheum Dis 2014, 73:1211–4. A and narrowband ultraviolet B therapy. Br J Dermatol 47. Baraliakos X, Baerlecken N, Witte T, Heldmann F, Braun 2010, 163:1282–90. J: High prevalence of anti-CD74 antibodies specific 60. Arican O, Aral M, Sasmaz S, Ciragil P: Serum levels for the HLA class II-associated invariant chain peptide of TNF-alpha, IFN-gamma, IL-6, IL-8, IL-12, IL-17, and (CLIP) in patients with axial spondyloarthritis. Ann IL-18 in patients with active psoriasis and correlation Rheum Dis 2014, 73:1079–82. with disease severity. Mediators Inflamm 2005, 48. Baeten D, Breban M, Lories R, Schett G, Sieper J: Are 2005:273–9. spondylarthritides related but distinct conditions or 61. Flisiak I, Klepacki a, Chodynicka B: Plasma and scales a single disease with a heterogeneous phenotype? levels of interleukin 18 in comparison with other Arthritis Rheum 2013, 65:12–20. possible clinical and laboratory biomarkers of psoriasis 49. Fredriksson T, Pettersson U: Severe psoriasis--oral activity. Biomarkers 2006, 11:194–200. therapy with a new retinoid. Dermatologica 1978, 62. Shimauchi T, Hirakawa S, Suzuki T, Yasuma A, 157:238–244. Majima Y, Tatsuno K, Yagi H, Ito T, Tokura Y: Serum 50. Beygi S, Lajevardi V, Abedini R: C-reactive protein interleukin-22 and vascular endothelial growth factor in psoriasis: a review of the literature. J Eur Acad serve as sensitive biomarkers but not as predictors Dermatol Venereol 2014, 28:700–11. of therapeutic response to biologics in patients with psoriasis. J Dermatol 2013, 40:805–12. 51. Coimbra S, Oliveira H, Reis F, Belo L, Rocha S, Quintanilha a., Figueiredo a., Teixeira F, Castro E, 63. Borská L, Fiala Z, Krejsek J, Andrýs C, Vokurková D, Rocha-Pereira P, Santos-Silva a.: C-reactive protein and Hamáková K, Ek JKČ, Ettler K: Selected Immunological leucocyte activation in psoriasis vulgaris according to Changes in Patients with sICAM-1 and IL-8.Physiol Res severity and therapy. J Eur Acad Dermatology Venereol 2006, 55:699–706. 2010, 24:789–796. 64. Elghandour TM, Youssef SES, Aly DG, Abd Elhameed 52. Nisa N, Qazi A: High-sensitivity C-reactive protein in MS, Abdel Moneim MM: Effect of Narrow Band psoriasis. Int J Dermatol 2012, 51:1393–1394. Ultraviolet B Therapy versus Methotrexate on Serum Levels of Interleukin-17 and Interleukin-23 in Egyptian 53. Gisondi P, Malerbai M, Malara G, Guerra AP, Sala R, Patients with Severe Psoriasis. Dermatol Res Pract Radaelp A, Girolomoni G, Civili S: C-reactive protein 2013, 2013:618269. and markers for thrombophilia in patients with chronic plaque psoriasis. Int J Immunopathol Pharmacol 2010, 65. Caproni M, Antiga E, Melani L, Volpi W, Del Bianco E, 23:1195–1202. Fabbri P: Serum levels of IL-17 and IL-22 are reduced by etanercept, but not by acitretin, in patients with 54. Strober B, Teller C, Yamauchi P, Miller JL, Hooper M, psoriasis: a randomized-controlled trial. J Clin Immunol Yang Y-C, Dann F: Effects of etanercept on C-reactive 2009, 29:210–4. protein levels in psoriasis and psoriatic arthritis. Br J

34 SERUM BIOMARKERS IN PSORIASIS AND IBD 66. Bonifati C, Trento E, Carducci M, Sacerdoti G, Mussi 79. Kamata M, Tada Y, Tatsuta a, Kawashima T, Shibata A, Fazio M, Ameglio F: Soluble E-selectin and soluble S, Mitsui H, Asano Y, Sugaya M, Kadono T, Kanda N, tumour necrosis factor receptor (60 kD) serum Watanabe S, Sato S: Serum lipocalin-2 levels are levels in patients with psoriasis. Dermatology 1995, increased in patients with psoriasis. Clin Exp Dermatol 190:128–131. 2012, 37:296–9. 67. Czech W, Schopf E, Kapp a.: Soluble E-selectin in 80. Romaní J, Caixàs A, Ceperuelo-Mallafré V, Carrascosa 2 sera of patients with atopic dermatitis and psoriasis- JM, Ribera M, Rigla M, Vendrell J, Luelmo J: Circulating correlation with disease activity. Br J Dermatol 1996, levels of lipocalin-2 and retinol-binding protein-4 are 134:17–21. increased in psoriatic patients and correlated with 68. Szepietowski J, Wa F, Bielicka E, Nockowski P, baseline PASI. Arch Dermatol Res 2013, 305:105–12. Noworolska A: Soluble E-selectin serum levels 81. Jansen P a M, Rodijk-Olthuis D, Hollox EJ, Kamsteeg correlate with disease activity in psoriatic patients.Exp M, Tjabringa GS, de Jongh GJ, van Vlijmen-Willems Dermatol 1999, 24:33–36. IMJJ, Bergboer JGM, van Rossum MM, de Jong EMGJ, 69. Krasowska D, Chodorowska G, Kozioł M, Czelej D: den Heijer M, Evers AWM, Bergers M, Armour J a L, Plasma levels of sICAM-1 in patients affected by Zeeuwen PLJM, Schalkwijk J: Beta-defensin-2 protein psoriasis : no relation to disease severity. Med Sci is a serum biomarker for disease activity in psoriasis Monit 2000, 6:353–355. and reaches biologically relevant concentrations in lesional skin. PLoS One 2009, 4:e4725. 70. Gangemi S, Merendino RA, Guarneri F, Minciullo PL, Dilorenzo G, Pacor M, Cannavò SP: Serum levels 82. Gambichler T, Bechara FG, Scola N, Rotterdam S, of interleukin-18 and s-ICAM-1 in patients affected Altmeyer P, Skrygan M: Serum levels of antimicrobial by psoriasis : preliminary considerations. Eur Acad peptides and proteins do not correlate with psoriasis Dermatology Venereol 2003:42–46. severity and are increased after treatment with fumaric acid esters. Arch Dermatol Res 2012, 304:471– 71. Long JW, Tao J, Pi X, Wang YY: Effect of Narrow-band 4. UVB Phototherapy on Soluble Cell Adhesion Molecules in Patients with Psoriasis Vulgaris. J Int Med reserach 83. Johnston a, Arnadottir S, Gudjonsson JE, Aphale a, 2014, 38:1507–1512. Sigmarsdottir a a, Gunnarsson SI, Steinsson JT, Elder JT, Valdimarsson H: Obesity in psoriasis: leptin and 72. Madej a, Reich a, Orda a, Szepietowski JC: Vascular resistin as mediators of cutaneous inflammation. Br J adhesion protein-1 (VAP-1) is overexpressed in Dermatol 2008, 159:342–50. psoriatic patients.J Eur Acad Dermatol Venereol 2007, 21:72–8. 84. Baran a, Flisiak I, Jaroszewicz J, Swiderska M: Serum adiponectin and leptin levels in psoriatic patients 73. Ataseven A, Kesli R: Novel inflammatory markers in according to topical treatment. J Dermatolog Treat psoriasis vulgaris: vaspin, vascular adhesion protein-1 2014, 6634:1–5. (vap-1), and ykl-40. G Ital Dermatol Venereol 2014. 85. Karadag AS, Ertugrul DT, Kalkan G, Bilgili SG, Celik HT, 74. Nemati H, Khodarahmi R, Rahmani A, Ebrahimi A, Takci Z, Balahoroglu R, Calka O: The effect of acitretin Amani M, Eftekhari K: Serum lipid profile in psoriatic treatment on insulin resistance, retinol-binding patients: correlation between vascular adhesion protein-4, leptin, and adiponectin in psoriasis vulgaris: Cell Biochem Funct protein 1 and lipoprotein (a). 2013, a noncontrolled study. Dermatology 2013, 227:103–8. 31(February 2012):36–40. 86. Oh YJ, Lim HK, Choi JH, Lee JW, Kim NI: Serum leptin 75. Schonthaler HB, Guinea-Viniegra J, Wculek SK, and adiponectin levels in Korean patients with Ruppen I, Ximénez-Embún P, Guío-Carrión A, Navarro psoriasis. J Korean Med Sci 2014, 29:729–34. R, Hogg N, Ashman K, Wagner EF: S100A8-S100A9 protein complex mediates psoriasis by regulating the 87. Zhu K-J, Zhang C, Li M, Zhu C-Y, Shi G, Fan Y-M: Leptin expression of complement factor C3. Immunity 2013, levels in patients with psoriasis: a meta-analysis. Clin 39:1171–81. Exp Dermatol 2013, 38:478–83. 76. Benoit S, Toksoy A, Ahlmann M, Schmidt M, 88. Shibata S, Saeki H, Tada Y, Karakawa M, Komine M, Sunderkotter C, Foell D, Pasparakis M, Roth J, Tamaki K: Serum high molecular weight adiponectin Goebeler M: Elevated serum levels of calcium-binding levels are decreased in psoriasis patients. J Dermatol S 100 proteins A 8 and A 9 reflect disease activity and Sci 2009, 55:62–3. abnormal differentiation of keratinocytes in psoriasis. 89. Ucak H, Demir B, Cicek D, Erden I, Aydin S, Dertlioglu Clin Lab Investig 2006, 7:62–66. SB, Arica M: Metabolic Changes and Serum Ghrelin 77. Ataseven A, Kesli R, Kurtipek GS, Ozturk P: Assessment Level in Patients with Psoriasis. Dermatol Res Pract of lipocalin 2, clusterin, soluble tumor necrosis factor 2014, 2014:1–6. receptor-1, interleukin-6, homocysteine, and uric acid 90. Ozdemir M, Yüksel M, Gökbel H, Okudan N, Mevlitoğlu levels in patients with psoriasis. Dis Markers 2014, I: Serum leptin, adiponectin, resistin and ghrelin 2014:541709. levels in psoriatic patients treated with cyclosporin. J 78. El-Hadidi H, Samir N, Shaker OG, Otb S: Estimation of Dermatol 2012, 39:443–8. tissue and serum lipocalin-2 in psoriasis vulgaris and 91. Kawashima K, Torii K, Furuhashi T, Saito C, Nishio its relation to metabolic syndrome.Arch Dermatol Res E, Nishida E, Shintani Y, Morita A: Phototherapy 2014, 306:239–45. reduces serum resistin levels in psoriasis patients.

CHAPTER 2 35 Photodermatol Photoimmunol Photomed 2011, 105. Van Praet L, Van den Bosch FE, Jacques P, Carron P, 27:152–5. Jans L, Colman R, Glorieus E, Peeters H, Mielants 92. Shibata S, Tada Y, Hau C, Tatsuta a, Yamamoto M, H, De Vos M, Cuvelier C, Elewaut D: Microscopic Kamata M, Karakawa M, Asano Y, Mitsui H, Sugaya gut inflammation in axial spondyloarthritis: a M, Kadono T, Saeki H, Kanda N, Sato S: Adiponectin multiparametric predictive model. Ann Rheum Dis as an anti-inflammatory factor in the pathogenesis 2013, 72:414–7. of psoriasis: induction of elevated serum adiponectin 106. Lees CW, Barrett JC, Parkes M, Satsangi J: New IBD levels following therapy. Br J Dermatol 2011, 164:667– genetics: common pathways with other diseases. Gut 70. 2011, 60:1739–53. 93. Bhushan M, McLaughlin B, Weiss JB, Griffiths CEM: 107. Harvey D, Pointon JJ, Evans DM, Karaderi T, Farrar C, Levels of endothelial cell stimulating angiogenesis Appleton LH, Sturrock RD, Stone M a, Oppermann factor and vascular endothelial growth factor are U, Brown M a, Wordsworth BP: Investigating the elevated in psoriasis. Br J Dermatol 1999, 141:1054– genetic association between ERAP1 and ankylosing 1060. spondylitis. Hum Mol Genet 2009, 18:4204–12. 94. Meki A-RM a, Al-Shobaili H: Serum vascular 108. Reveille JD, Sims A-M, Danoy P, Evans DM, Leo P, endothelial growth factor, transforming growth factor Pointon JJ, Jin R, Zhou X, Bradbury L a, Appleton LH, β1, and nitric oxide levels in patients with psoriasis Davis JC, Diekman L, Doan T, Dowling A, Duan R, Duncan vulgaris: their correlation to disease severity.J Clin Lab EL, Farrar C, Hadler J, Harvey D, Karaderi T, Mogg R, Anal 2014, 28:496–501. Pomeroy E, Pryce K, Taylor J, Savage L, Deloukas P, 95. Flisiak I, Zaniewski P, Rogalska-Taranta M, Chodynicka Kumanduri V, Peltonen L, Ring SM, Whittaker P, et B: Effect of psoriasis therapy on VEGF and its soluble al.: Genome-wide association study of ankylosing receptors serum concentrations. J Eur Acad Dermatol spondylitis identifies non-MHC susceptibility loci. Nat Venereol 2012, 26:302–7. Genet 2010, 42:123–7. 96. Tekin NS, Ilter N, Sancak B, Ozden MG, Gurer MA: 109. Tremelling M, Cummings F, Fisher S a., Mansfield Nitric Oxide Levels in Patients with Psoriasis Treated J, Gwilliam R, Keniry A, Nimmo ER, Drummond H, with Methotrexate. Mediators Inflamm 2006, 2006:1– Onnie CM, Prescott NJ, Sanderson J, Bredin F, Berzuini 5. C, Forbes A, Lewis CM, Cardon L, Deloukas P, Jewell D, Mathew CG, Parkes M, Satsangi J: IL23R Variation 97. Gokhale N, Belgaumkar V, Pandit D, Deshpande Determines Susceptibility But Not Disease Phenotype S, Damle D: A study of serum nitric oxide levels in in Inflammatory Bowel Disease. Gastroenterology Indian J Dermatol Venereol Leprol psoriasis. 2005, 2007, 132:1657–1664. 71:175. 110. Bernstein CN, Fried M, Krabshuis JH, Cohen H, Eliakim 98. Zalewska A, Wyczółkowska J, Narbutt J, Sysa- R, Fedail S, Gearry R, Goh KL, Hamid S, Khan AG, Jędrzejowska A: Nitric oxide levels in plasma and LeMair a W, Malfertheiner, Ouyang Q, Rey JF, Sood A, J fibroblast cultures of psoriasis vulgaris patients. Steinwurz F, Thomsen OO, Thomson A, Watermeyer Dermatol Sci 2007, 48:237–240. G: World Gastroenterology Organization Practice 99. Serwin AB, Sokolowska M, Chodynicka B: Tumour Guidelines for the diagnosis and management of IBD necrosis factor a (TNF- alpha)-converting enzyme in 2010. Inflamm Bowel Dis 2010, 16:112–24. (TACE) and soluble TNF- a receptor type 1 in psoriasis 111. Best W, Becktel J, Singleton J, Jr. KF: Development of patients treated with narrowband ultraviolet B. a Crohn’s disease activity index. National Cooperative Photodermatol Photoimmunol Photomed 2007, Crohn’s Disease Study. Gastroenterology 1976, 23:130–134. 70:439–444. 100. Dekker-saeys BJ, Meuwissen SGM, Den EMVAN: IL . 112. Walmsley RS, Ayres RC, Pounder RE, Allan RN: A simple Prevalence of peripheral arthritis , sacroiliitis , and clinical colitis activity index. Gut 1998, 43:29–32. ankylosing spondylitis in patients suffering from inflammatory bowel disease. 1978:33–35. 113. Schroeder K, Tremaine W, Ilstrup D: Coated Oral 5-Aminosalicylic Acid Therapy for Mildly to Moderately 101. Orchard TR, Wordsworth BP, Jewell DP: Peripheral Active Ulcerative Colitis.N Engl J Med 1987, 317:1625– arthropathies in inflammatory bowel disease: their 1629. articular distribution and natural history. Gut 1998, 42:387–391. 114. Poullis AP, Zar S, Sundaram KK, Moodie SJ, Risley P, Theodossi A, Mendall MA: A new , highly sensitive 102. Palm O, Moum B, Ongre A, Gran JT, Palm Ø, Gran assay for C-reactive protein can aid the differentiation JANT: Patients with inflammatory bowel disease : a of in ¯ ammatory bowel disorders from constipation- population study (the IBSEN study). 2002, 29. and diarrhoea-predominant functional bowel 103. Salvarani C: Clinical features and epidemiology of disorders. Eur J Gastroenterol Hepatol 2002, 14:409– spondyloarthritides associated with inflammatory 412. World J Gastroenterol bowel disease. 2009, 15:2449. 115. Fagan E, Dyck R, Maton P, Hodgson H, Chadwick V, 104. De Vos M, Cuvelier C, Mielants H, Veys E, Barbier F, A. E: Petrie A, Pepys M: Serum levels of C-reactive protein in Ileocolonoscopy in seronegative spondylarthropathy. Crohn’s disease and ulcerative colitis. Eur J Clin Invest Gastroenterology 1989:339–344. 1982, 12:351–359.

36 SERUM BIOMARKERS IN PSORIASIS AND IBD 116. Chamouard P, Richert Z, Meyer N, Rahmi G, Baumann clinical activity in iflammatory bowel disease. J Clin R: Diagnostic value of C-reactive protein for predicting Gastroenterol 1986, 8:647–650. activity level of Crohn’s disease. Clin Gastroenterol 129. Balding J, Livingstone WJ, Conroy J, Mynett-Johnson Hepatol 2006, 4:882–7. L, Weir DG, Mahmud N, Smith OP: Inflammatory 117. Jürgens M, Mahachie John JM, Cleynen I, Schnitzler bowel disease: the role of inflammatory cytokine gene F, Fidder H, van Moerkercke W, Ballet V, Noman M, polymorphisms. Mediators Inflamm 2004, 13:181–7. 2 Hoffman I, van Assche G, Rutgeerts PJ, van Steen K, 130. Shanahan F: Crohn’s disease. Lancet 2002, 359:62–9. Vermeire S: Levels of C-reactive protein are associated with response to infliximab therapy in patients with 131. Gross V, Andus T, Caesar I, Roth M: Evidence for Crohn’s disease. Clin Gastroenterol Hepatol 2011, continuous stimulation of interleukin-6 production in 9:421–7.e1. Crohn ’s disease. Gastroenterology 1992, 102:514– 519. 118. Louis E, Vermeire S, Rutgeerts P, Vos M De, Gossum A Van, Pescatore P, Fiasse R, Pelckmans P, Reynaert H, 132. Hyams JS, Fitzgerald JE, Treem WR, Wyzga N, Kreutzer Haens GD, Malaise M, Belaiche J, Louis E, Gastroente S DL: Relationship of functional and antigenic interleukin De: A Positive Response to infliximab in crohn disease: 6 to disease activity in inflammatory bowel disease. Association with a higher systemic inflammation Gastroenterology 1993, 104:1285–1292. before treatment but not with - 308 TNF Gene 133. Nancey S, Hamzaoui N, Moussata D, Graber I, Bienvenu Polymorphism. Scand J Clin Lab Invest 2002, 7:818– J, Flourie B: Serum interleukin-6, soluble interleukin-6 824. receptor and Crohn’s disease activity.Dig Dis Sci 2008, 119. Solem C a, Loftus E V, Tremaine WJ, Harmsen 53:242–247. WS, Zinsmeister AR, Sandborn WJ: Correlation 134. Goke M, Hoffmann JC, Evers J, Kroger H, Manns MP: of C-Reactive Protein With Clinical, Endoscopic, Elevated serum concentrations of soluble selectin and Histologic, and Radiographic Activity in Inflammatory immunoglobulin type adhesion molecules in patients Bowel Disease. Inflamm Bowel Dis 2005, 11:707–712. with inflammatory bowel disease. J Gastroenterol 120. Wright J, Alp M, Young F, Tigler-Wybrandi N: Predictors 1997, 32:480–486. of Acute Relapse of Crohn ’ s Disease A Laboratory and 135. Jones SC, Banks RE, Haidar a, Gearing a J, Hemingway Clinical Study. Dig Dis Sci 1987, 32:164–170. IK, Ibbotson SH, Dixon MF, Axon a T: Adhesion 121. Gustot T, Lemmers a, Louis E, Nicaise C, Quertinmont molecules in inflammatory bowel disease. Gut 1995, E, Belaiche J, Roland S, Van Gossum a, Devière J, 36:724–730. Franchimont D: Profile of soluble cytokine receptors in 136. Magro F, Araujo F, Pereira P, Meireles E: Soluble Crohn’s disease. Gut 2005, 54:488–95. Selectins , sICAM , sVCAM , and Angiogenic 122. Koutroubakis IE, Petinaki E, Dimoulios P, Vardas E, Proteins in Different Activity Groups of Patients Roussomoustakaki M, Maniatis AN, Kouroumalis E with Inflammatory Bowel Disease. Dig Dis Sci 2004, a: Increased serum levels of YKL-40 in patients with 49(August):1265–1274. inflammatory bowel disease. Int J Colorectal Dis 2003, 137. Patel R, Abeed A, Dwomoa A, Keighley M: Circulating 18:254–9. soluble adhesion molecules in inflammatory bowel 123. Vind I, Johansen J, Price P, Munkholm P: Serum YKL- disease. Eur J Gastroenterol Hepatol 1995. 40, a Potential New Marker of Disease Activity in 138. Song W-B: Soluble intercellular adhesion molecule-1, Patients with Inflammatory Bowel Disease. Scand J D-lactate and diamine oxidase in patients with Gastroenterol 2003, 38:599–605. inflammatory bowel disease. World J Gastroenterol 124. Erzin Y, Uzun H, Karatas A, Celik AF: Serum YKL-40 as 2009, 15:3916. a marker of disease activity and stricture formation in 139. Ates Y, Degertekin B, Erdil A, Yaman H, Dagalp K: Serum patients with Crohn’s disease. J Gastroenterol Hepatol ghrelin levels in inflammatory bowel disease with 2008, 23(8 Pt 2):e357–62. relation to disease activity and nutritional status. Dig 125. Herrlinger KR, Dittmann R, Weitz G, Wehkamp J, Dis Sci 2008, 53:2215–21. Ludwig D, Schwab M, Stange EF, Fellermann K: Serum 140. Peracchi M, Bardella MT, Caprioli F, Massironi S, Conte Procalcitonin Differentiates Inflammatory Bowel D, Valenti L, Ronchi C, Beck-Peccoz P, Arosio M, Piodi L: Disease and Self-limited Colitis. Inflamm Bowel Dis Circulating ghrelin levels in patients with inflammatory 2004, 10:229–233. bowel disease. Gut 2006, 55:432–3. 126. Oruç N, Z ÖÖ, Lu NO, Lter T: Diagnostic value of serum 141. Sung EZH, Da Silva NF, Goodyear S, McTernan PG, procalcitonin in determining. Turkish J Gastroenterol Sanger GJ, Nwokolo CU: Increased plasma ghrelin 2009, 20:9–12. following infliximab in Crohn’s disease. Aliment 127. Oussalah a, Laurent V, Bruot O, Guéant J-L, Régent Pharmacol Ther 2009, 29:83–9. D, Bigard M, Peyrin-Biroulet L: Additional benefit of 142. Kader N, El-Din F, Khatab E, Elabd N: Does plasma procalcitonin to C-reactive protein to assess disease resistin level have a role in predicting inflammatory activity and severity in Crohn’s disease. Aliment bowel disease activity? Indian J Gastroenterol 2010, Pharmacol Ther 2010, 32:1135–44. 29:126–127. 128. Sachar D, Smith H, Chan S, Cohen L, Lichtiger S: 143. Konrad A, Lehrke M, Schachinger V, Seibold F, Stark Erythrocytic Sedimentation Rate as a measure of R, Ochsenkuhn T, Parhofer KG, Goke B, Broedl UC:

CHAPTER 2 37 Resistin is an inflammatory marker of inflammatory 155. Leach ST, Yang Z, Messina I, Song C, Geczy CL, bowel disease in humans. Eur J Gastroenterol Hepatol Cunningham AM, Day AS: Serum and mucosal S100 2007, 19:1070–1074. proteins, calprotectin (S100A8/S100A9) and S100A12, 144. Morisaki T, Takeshima F, Fukuda H, Matsushima are elevated at diagnosis in children with inflammatory K, Akazawa Y, Yamaguchi N, Ohnita K, Isomoto H, bowel disease. Scand J Gastroenterol 2007, 42:1321– Takeshita H, Sawai T, Fujita F, Nakao K: High serum 31. vaspin concentrations in patients with ulcerative 156. Malícková K, Kalousová M, Fucíková T, Bortlík M, colitis. Dig Dis Sci 2014, 59:315–21. Duricová D, Komárek V, Zima T, Janatková I, Lukás 145. Rodrigues VS, Milanski M, Fagundes JJ, Torsoni a M: Anti-inflammatory effect of biological treatment S, Ayrizono MLS, Nunez CEC, Dias CB, Meirelles LR, in patients with inflammatory bowel diseases: Dalal S, Coy CSR, Velloso L a, Leal RF: Serum levels calprotectin and IL-6 changes do not correspond to and mesenteric fat tissue expression of adiponectin sRAGE changes. Scand J Clin Lab Invest 2010, 70:294– and leptin in patients with Crohn’s disease. Clin Exp 9. Immunol 2012, 170:358–64. 157. Von Roon AC, Karamountzos L, Purkayastha S, Reese 146. Karmiris K, Koutroubakis IE, Xidakis C, Polychronaki M, GE, Darzi AW, Teare JP, Paraskeva P, Tekkis PP: Diagnostic Voudouri T, Kouroumalis EA: Circulating levels of leptin, precision of fecal calprotectin for inflammatory bowel adiponectin, resistin, and ghrelin in inflammatory disease and colorectal malignancy. Am J Gastroenterol bowel disease. Inflamm Bowel Dis 2006, 12:100–105. 2007, 102:803–13. 147. Weigert J, Obermeier F, Neumeier M, Wanninger 158. Van Rheenen PF, Van de Vijver E, Fidler V: Faecal J, Filarsky M, Bauer S, Aslanidis C, Rogler G, Ott C, calprotectin for screening of patients with suspected Schaffler A, Scholmerich J, Buechler C: Circulating inflammatory bowel disease: diagnostic meta-analysis. levels of chemerin and adiponectin are higher in BMJ 2010, 341:c3369. ulcerative colitis and chemerin is elevated in Crohn’s 159. Lin J-F, Chen J-M, Zuo J-H, Yu A, Xiao Z-J, Deng F-H, disease. Inflamm Bowel Dis 2010, 16:630–637. Nie B, Jiang B: Meta-analysis: fecal calprotectin for 148. Franchimont D, Roland S, Gustot T, Quertinmont E, assessment of inflammatory bowel disease activity. Toubouti Y, Gervy M-C, Deviere J, Van Gossum A: Inflamm Bowel Dis 2014, 20:1407–15. Impact of infliximab on serum leptin levels in patients 160. Wagner M: Fecal markers of inflammation used as with Crohn’s disease. J Clin Endocrinol Metab 2005, surrogate markers for treatment outcome in relapsing 90:3510–6. inflammatory bowel disease. World J Gastroenterol 149. Karmiris K, Koutroubakis IE, Xidakis C, Polychronaki M, 2008, 14:5584. Kouroumalis EA: The effect of infliximab on circulating 161. Sipponen T, Björkesten C-G a F, Färkkilä M, Nuutinen levels of leptin, adiponectin and resistin in patients H, Savilahti E, Kolho K-L: Faecal calprotectin with inflammatory bowel disease. Eur J Gastroenterol and lactoferrin are reliable surrogate markers Hepatol 2007, 19:789–794. of endoscopic response during Crohn’s disease 150. Manolakis AC, Kapsoritakis AN, Georgoulias P, treatment. Scand J Gastroenterol 2010, 45:325–31. Tzavara C, Valotassiou V, Kapsoritaki A, Potamianos 162. Kolho K-L, Sipponen T: The long-term outcome of SP: Moderate performance of serum S100A12, in anti-tumor necrosis factor-α therapy related to distinguishing inflammatory bowel disease from fecal calprotectin values during induction therapy irritable bowel syndrome. BMC Gastroenterol 2010, in pediatric inflammatory bowel disease. Scand J 10:118. Gastroenterol 2014, 49:434–41. 151. Brinar M, Cleynen I, Coopmans T, Van Assche G, 163. Hämäläinen A, Sipponen T, Kolho K-L: Infliximab Rutgeerts P, Vermeire S: Serum S100A12 as a new in pediatric inflammatory bowel disease rapidly marker for inflammatory bowel disease and its decreases fecal calprotectin levels. World J relationship with disease activity. Gut 2010:1728– Gastroenterol 2011, 17:5166–71. 1730. 164. Ferreiro-iglesias R, Acosta MB, Santiago MO, Gonzalez 152. Schimid KW, Lugering N, Stoll R, Brinkb P, Winde G, L, Alonso C, Pen D, Estevez JB, Dominguez-mun Domschke W, Bocker W, Sorg C: Immunohistochemical JE: Fecal Calprotectin as Predictor of Relapse in Demonstration of the Calcium-Binding Proteins MRP8 Patients With Inflammatory Bowel Disease Under and MRP14 and Their Heterodimer (27E 10 Antigen) in Maintenance. J Clin Gastroenterol 2015;00:000–000. Crohn ’ s Disease. Hum Pathol 1995, 26:334–337. 165. Wright EK, Kamm M a, De Cruz P, Hamilton AL, Ritchie 153. Lugering N, Stoll R, Schmid K, Kucharzik T, Stein H, KJ, Krejany EO, Leach S, Gorelik A, Liew D, Prideaux L, Burmeisterj G: The myeloic related protein MRP8/14 Lawrance IC, Andrews JM, Bampton P a, Jakobovits SL, (27E10 antigen) - usefulness as a potential marker Florin TH, Gibson PR, Debinski H, Macrae F a, Samuel for disease activity in ulcerative colitis and putative D, Kronborg I, Radford-Smith G, Selby W, Johnston biological function. Hum Pathol 1995, 14:659–664. MJ, Woods R, Elliott PR, Bell SJ, Brown SJ, Connell 154. Meuwis M, Vernier-Massouille G, Grimaud JC, Bouhnik WR, Day AS, Desmond P V, et al.: Measurement of Y, Laharie D, Piver E, Seidel L, Colombel JF, Louis E: fecal calprotectin improves monitoring and detection Serum calprotectin as a biomarker for Crohn’s disease. of recurrence of Crohn’s disease after surgery. J Crohns Colitis 2013, 7:e678–83. Gastroenterology 2015, 148:938–947.e1.

38 SERUM BIOMARKERS IN PSORIASIS AND IBD 166. Janas RM, Ochocińska A, Snitko R, Dudka D, Kierkuś Zinsmeister AR, Targan SR: Evaluation of Serologic J, Teisseyre M, Najberg E: Neutrophil gelatinase- Disease Markers in a Population-Based Cohort of associated lipocalin in blood in children with Patients with Ulcerative Colitis and Crohn’s Disease. inflammatory bowel disease. J Gastroenterol Hepatol Inflamm Bowel Dis 2001, 7:192–201. 2014, 29:1883–9. 181. Esters N, Vermeire S, Joossens S, Noman M, Louis 167. Oikonomou KA, Kapsoritakis AN, Theodoridou C, E, Belaiche J, De Vos M, Van Gossum a, Pescatore P, 2 Karangelis D, Germenis A, Stefanidis I, Potamianos Fiasse R, Pelckmans P, Reynaert H, Poulain D, Bossuyt SP: Neutrophil gelatinase-associated lipocalin X, Rutgeerts P: Serological markers for prediction of (NGAL) in inflammatory bowel disease: association response to anti-tumor necrosis factor treatment in with pathophysiology of inflammation, established Crohn’s disease. Am J Gastroenterol 2002, 97:1458– markers, and disease activity. J Gastroenterol 2012, 62. 47:519–530. 182. Smith BRK, Arnott IDR, Drummond HE, Nimmo ER: 168. Yeşil A, Gönen C, Senateş E, Paker N, Gökden Y, Disease location, anti-saccharomyces cerevisiae Koçhan K, Erdem ED, Gündüz F: Relationship between Antibody, and NOD2/CARD15 genotype influence the neutrophil gelatinase-associated lipocalin (NGAL) progression of disease behavior in crohn’s disease. levels and inflammatory bowel disease type and Inflamm Bowel Dis 2004, 10:521–528. activity. Dig Dis Sci 2013, 58:2587–93. 183. Bonneau J, Dumestre-Perard C, Rinaudo-Gaujous 169. Kim JM: Antimicrobial Proteins in Intestine and M, Genin C, Sparrow M, Roblin X, Paul S: Systematic Inflammatory Bowel Diseases. Intest Res 2014, review: New serological markers (anti-glycan, anti- 9100:20–33. GP2, anti-GM-CSF Ab) in the prediction of IBD patient 170. Yamaguchi N, Isomoto H, Mukae H, Ishimoto H, outcomes. Autoimmun Rev 2014, 14:231–245. Ohnita K, Shikuwa S, Mizuta Y, Nakazato M, Kohno S: 184. Karmiris K, Koutroubakis IE, Kouroumalis E a: Leptin, Concentrations of alpha- and beta-defensins in plasma adiponectin, resistin, and ghrelin--implications for of patients with inflammatory bowel disease.Inflamm inflammatory bowel disease.Mol Nutr Food Res 2008, Res 2009, 58:192–7. 52:855–66. 171. Zissis M, Afroudakis A, Galanopoulos G, Palermos I, 185. Ates Y, Degertekin B, Erdil A, Yaman H, Dagalp K: Serum Boura X, Sc B, Michopoulos S, Archimandritis A: B2 ghrelin levels in inflammatory bowel disease with Microglobulin: Is It a Reliable Marker of Activity in relation to disease activity and nutritional status. Dig Inflammatory Bowel Disease? 2001, 96. Dis Sci 2008, 53:2215–2221. 172. Yılmaz B, Köklü S, Yüksel O, Arslan S: Serum beta 186. Sari I, Kebapcilar L, Taylan A, Bilgir O, Kozaci DL, 2-microglobulin as a biomarker in inflammatory bowel Yildiz Y, Yuksel A, Gunay N, Akkoc N: Fetuin-A and disease. World J Gastroenterol 2014, 20:10916–20. interleukin-18 levels in ankylosing spondylitis. Int J 173. Dotan I: New serologic markers for inflammatory Rheum Dis 2010, 13:75–81. bowel disease diagnosis. Dig Dis 2010, 28:418–23. 187. Wendling D, Racadot E, Auge B, Toussirot E: 174. Homsak E, Micetić-Turk D, Bozic B: Autoantibodies Soluble intercelulular adhesion molecule 1 in pANCA, GAB and PAB in inflammatory bowel disease: spondyloarthritis. Clin Rheumatol 1998, 17:202–204. prevalence, characteristics and diagnostic value. Wien 188. Kemeny-Beke A, Gesztelyi R, Bodnar N, Zsuga J, Klin Wochenschr 2010, 122 Suppl :19–25. Kerekes G, Zsuga M, Biri B, Keki S, Szodoray P, Betera A. 175. Kuna AT: Review Serological markers of inflammatory Increased production of asymmetric dimethylarginine bowel disease. Biochem Medica 2013, 23:28–42. (ADMA) and atherosclerotic disease in ankylosing spondylitis. Joint bone spine 2011, 78:184–187. 176. Bossuyt X: Serologic markers in inflammatory bowel disease. Clin Chem 2006, 52:171–81. 189. Erre G, Sanna P, Zinellu A, Ponchietti A, Fenu P, Sotgia S, Carru C, Gannau A, Passiu G: Plasma asymmetric 177. Prideaux L, De Cruz P, Ng SC, Kamm M a: Serological dimethylarginine (ADMA) levels and atherosclerotic antibodies in inflammatory bowel disease: a disease in akylosing spondylitis: A cross-sectional Inflamm Bowel Dis systematic review. 2012, 18:1340– study. Clin Rheumatol 2011, 30:21–27. 55. 190. O’Shea F, Tsui F, Chiu B, Tsui H, Yazdanpanah M, Innman 178. Peeters M, Ph D, Joossens S, Sc M, Vermeire S, R: Retinol (vitamin A) and retinol binding protein levels Vlietinck R, Bossuyt X, Rutgeerts P: Diagnostic Value are decreased in ankylosing spondylitis: clinical and of Anti- Saccharomyces cerevisiae and Antineutrophil genetic analysis. J Rheumatol 2007, 34:2457–2459. Cytoplasmic Autoantibodies in Inflammatory Bowel Disease. Am J Gastroenterol 2001, 96:731–734. 191. Genre F, Lopez-Mejias R, Miranda-Filloy J, Ubilla B, Carnero-Lopez B, Gomez-Acebo I, Blanco R, Ochoa 179. Ruemmele FM, Targan SR, Levy G, Dubinsky M, Braun R, Rueda-Gotor J, Pina T, Gozalez-Juanatay C, Llorca J, Seidman EG: Diagnostic accuracy of serological J, Gonzalez-Gay M: Antitumour necrosis factor alpha assays in pediatric inflammatory bowel disease. treatment reduces retinol-binding protein 4 serum Gastroenterology 1998, 115:822–829. levels in non-dianetic ankylosing spondylitis patients. 180. Sandborn WJ, Loftus E V., Colombel JF, Fleming K Ann Rheum Dis 2014, 73:941–943. a., Seibold F, Homburger H a., Sendid B, Chapman 192. Ozgocmen S, Sogus S, Ardicoglu O, Fadillioglu E, I RW, Tremaine WJ, Kaul DK, Wallace J, Harmsen WS, P, Akyol O: Serum nitric oxide, catalase, superoxide

CHAPTER 2 39 dismtase and malondialdehyde stats in patients with stress and the unfolded protein response are linked to ankylsing spondylitis. Rheumatol Int 2004, 24:80–83. synergistic IFN-β induction via X-box binding protein. 193. Kozaci L, Sari I, Alacacioglu A, Akar S, Akkoc N: Eur J immmunology 2008, 38:1194–203. Evaluation of inflammation and oxidative stress 205. Van Duivenvoorde LM, Dorris ML, Satumtira N, in ankylosing spondylitis: a role for macrophage van Tok MN, Redlich K, Tak PP, Taurog JD, Baeten migration inhibitory factor. Mod Rheumatol 2010, DL: Relationship between inflammation, bone 20:34–39. destruction, and osteoproliferation in the HLA-B27/ 194. Hoffman I, Dmetter P, Peeters M, De Vos M, Mielants human β2 -microglobulin-transgenic rat model of H, Veys E, De Keyser F: Anti-saccharomyces cerevisiae spondylarthritis. Arthritis Rheum 2012, 64:3210–9. IgA antibodies are raised in ankylosing spondylitis and 206. Tran TM, Dorris ML, Satumtira N, Richardson J a, undifferentiated spondyloarthropathy.Ann Rheum Dis Hammer RE, Shang J, Taurog JD: Additional human 2003, 62:455–459. beta2-microglobulin curbs HLA-B27 misfolding and 195. Aydin S, Atugunduz P, Temel M, Bicakcigil M, Tasan promotes arthritis and spondylitis without colitis in D, Direskeneli H: Anti-saccharomyces cerevisiae male HLA-B27-transgenic rats. Arthritis Rheum 2006, antibodies (ASCA) in spondyloarthropathies: a 54:1317–27. reassessment. Rheumatology (Oxford) 2008, 47:142– 207. Hammer RE, Maika SD, Richardson JA, Tang J, Taurog 144. JD: Spontaneous Inflammatory Disease in Transgenic 196. De Vries M, van der Horst-Bruinsma I, van Hoogstraten Rats Expressing HLA-B27 and Human Pzm : An Animal I, van Bodegraven A, von Blomberg BME, Ratnawati Model of HLRB27-Associated Human Disorders. Cell H, Dijkmans B: pANCA, ASCA, and OmpC antibodies 1990, 63:1099–1112.7 in patients with ankylosing spondylitis without inflammatory bowel disease. J Rheumatol 2010, 37:2340–2344. 197. Appel H, Maier R, Wu P, Scheer R, Hempfing A, Kayser R, Thiel A, Radbruch A, Loddenkemper C, Sieper J: Analysis of IL-17(+) cells in facet joints of patients with spondyloarthritis suggests that the innate immune pathway might be of greater relevance than the Th17- mediated adaptive immune response. Arthritis Res Ther 2011, 13:R95. 198. Di Meglio P, Di Cesare A, Laggner U, Chu C-C, Napolitano L, Villanova F, Tosi I, Capon F, Trembath RC, Peris K, Nestle FO: The IL23R R381Q gene variant protects against immune-mediated diseases by impairing IL-23-induced Th17 effector response in humans. PLoS One 2011, 6:e17160. 199. Sarin R, Wu X, Abraham C: Inflammatory disease protective R381Q IL23 receptor polymorphism results in decreased primary CD4 + and CD8 + human T-cell functional responses. Proc Natl Acad Sci 2011, 108:9560–9565. 200. Raychaudhuri SK, Saxena A, Raychaudhuri SP: Role of IL-17 in the pathogenesis of psoriatic arthritis and axial spondyloarthritis. Clin Rheumatol 2015, 34:1019–23. 201. DeLay M, Turner M, Klenk E: HLA-B27 misfolding and the unfolded protein response augment interleukin-23 production and are associated with Th17 activation in transgenic rats. Arthritis Rheum 2009, 60:2633–2643. 202. Zhang L, Li Y, Li Y, Qi L, Liu X, Yuan C, Hu N, Ma D, Li Z, Yang Q, Li W, Li J: Increased frequencies of Th22 cells as well as Th17 cells in the peripheral blood of patients with ankylosing spondylitis and rheumatoid arthritis. PLoS One 2012, 7:e31000. 203. Kollnberger S, Bird L, Roddis M: HLA-B27 heavy chain homodimers are expressed in HLA-B27 transgenic rodent models of spondyloarthritis and are ligands for paired Ig-like receptors. J Immunol 2004, 173:1699– 710. 204. Smith J, Turner M, DeLay M: Endoplasmic reticulum

40 SERUM BIOMARKERS IN PSORIASIS AND IBD 2

CHAPTER 2 41

Serum inflammatory biomarkers fail to identify early axial spondyloartritis: results from the SpondyloArthritis Caught Early (SPACE)-cohort

Maureen C. Turina1, Nataliya Yeremenko1,2, Floris van Gaalen3, Maikel van Oosterhout4, Inger J. Berg5, Ramona Ramonda6, Maria C. Lebre2, Robert Landewé1, Dominique L. Baeten1,2

1Amsterdam Rheumatology and Immunology Center, Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and immunology Center, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; 2Department of Experimental Immunology, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; 3Department of Rheumatology, Leiden University Medical Center, Leiden, The Netherlands; 4 3 Department of Rheumatology, Groene Hart Ziekenhuis, Gouda , The Netherlands; 5 Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway; 6 Rheumatology Unit, Department of Medicine- DIMED- University of Padova, Padova Italy

Submitted for publication ABSTRACT

Introduction Decreasing the diagnostic delay in axial spondyloarthritis (axSpA) remains a major challenge. Here, we assessed the value of serum inflammatory biomarkers to distinguish early axSpA from other pathologies in a large cohort of patients referred with early back pain.

Methods Serum c-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and calprotectin were determined in the SPondyloArthritis Caught Early (SPACE) cohort (n=310), an early back pain inception cohort. Additionally, explorative serum biomarkers derived from literature (interleukin-27 [IL-27], human beta-defensin-2 [hBD-2], and lipocalin-2 [LCN-2]) were determined by ELISA in full-blown ankylosing spondylitis (AS) patients (n=21) and healthy controls (n=19).

Results Serum CRP and ESR levels were not elevated in early axSpA versus ‘control’ back pain patients. Serum calprotectin was elevated in early axSpA versus controls (p=0.01) but failed to identify early axSpA at the individual level (positive predictive value of 38.7%). Asto explorative biomarkers, serum levels of IL-27 were not detectable, and hBD-2 and LCN-2 serum levels were not elevated in full-blown AS versus healthy controls (p=0.572, p=0.562, respectively). Therefore, these markers were not further determined in the SPACE cohort.

Conclusion None of the candidate serum inflammatory markers were useful as diagnostic markers in the early phase of axSpA.

44 SERUM INFLAMMATORY MARKERS NOT IN EARLY SpA INTRODUCTION

Reliable diagnosis of axial spondyloarthritis (axSpA) in the earliest phases of the disease remains an important unmet medical need. Ample evidence indicates that 1) signs and symptoms of active disease are as severe in early disease as in full-blown disease, with similar impact on function and quality of life,1–3 and 2) active early axSpA can be effectively treated 3 with non-steroidal anti-inflammatory drugs (NSAIDs) and/or tumor necrosis factor inhibitors (TNFi).4–6 Moreover, current treatments fail to significantly inhibit pathological new bone formation when started in full-blown disease;7–9 although awaiting formal proof, starting the same treatments in very early disease may also impact structural progression.10,11 Although the time between first symptoms and diagnosis of axSpA has already been significantly reduced over the last decade by a combination of early referral strategies and the use of MRI to image axial inflammation,12–16 it remains a challenge to further reduce the diagnostic delay and to reliably distinguish early back pain due to axSpA from other causes of back pain.17–19

One potential way to address this challenge is the use of serum inflammatory biomarkers. A couple of inflammatory biomarkers have been reported to be elevated in active full-blown axSpA. C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) are both acute- phase reactants which are elevated in active SpA and decrease upon effective treatment.20–23 Moreover, calprotectin (also called S100A8/A9), a calcium binding protein, is expressed and secreted during macrophage infiltration in SpA synovitis.24–26 Calprotectin was recently shown to be a good serum biomarker for treatment responses in proof-of-concept trials in SpA and to independently predict radiographic progression in axSpA.27,28 Despite their value at the group level, the value of these serum biomarkers for diagnosis of axSpA in patients with early back pain remains unknown as they are neither very sensitive (for example, only one third of patients with active ankylosing spondylitis (AS) have an elevated CRP)20,22 nor specific as they reflect inflammation (whatever the origin) rather than axSpA as such.

Besides these inflammatory biomarkers that were already extensively studied in AS, our recent literature review identified a few other potentially interesting serum biomarkers (M. Turina et al., submitted for publication). Lin et al.29 recently described that levels of interleukin-27 (IL-27), a heterodimeric cytokine composed of p28 and Epstein–Barr Virus-induced gene 3 (EBI3) which belongs to the IL-12 family, were elevated in full-blown AS when compared with healthy controls but this finding has not yet been confirmed in an independent study. Human beta defensin-2 (hBD-2) and lipocalin-2 (LCN-2) are two antimicrobial peptides which are up-regulated by IL-17 and are consistently found at elevated levels in serum of patients with active psoriasis and inflammatory bowel disease (IBD).30–35 Considering the central role of IL-17 in the pathophysiology of axSpA36–38 and the pathophysiological and clinical overlap of axSpA with psoriasis and IBD,39 these two biomarkers could also be of potential interest.

CHAPTER 3 45 Accordingly, the aim of the present study was to assess the biomarker value of established inflammatory serum markers (CRP, ESR, and calprotectin) as well as more exploratory biomarkers (IL-27, hBD-2, and LCN-2) for the diagnosis of axSpA in patients with early back pain.

METHODS

Patients and samples Serum was collected from 350 individuals after obtaining written informed consent to participate to the studies as approved by the local Ethics Committees of the participating centers. Cohort 1 (SPondyloArthritis Caught Early [SPACE]) consisted of 310 individuals with ‘early’ back pain (defined as at least 3 and maximally 24 months) and an age of onset below 45 years, referred for early evaluation of potential AxSpA.40 Cohort 2 consisted of 21 established AS patients according to the modified New York (mNY) criteria; all patients were TNFi naïve.41 Full descriptions of these cohorts were reported previously.40,41 Finally, we also obtained serum from 19 healthy controls.

Serum inflammatory biomarkers From cohort 1, serum CRP and ESR levels were determined by local laboratories. Serum calprotectin levels were determined by enzyme-linked immunosorbent assay (ELISA) (Hycult Biotech, Uden, the Netherlands) using a 1:60 dilution and according to the manufacturer’s protocol.27,28

From cohort 2 and the healthy controls, serum IL-27, hBD-2, and LCN-2 levels were determined by ELISA according to the manufacturer’s protocol (eBiosciences, San Diego, CA; Phoenix pharmaceuticals, Inc. Belmont CA; and Research & Diagnostic Systems, Inc. Minneapolis, respectively). The dilutions were 1:2, 1:50, and 1:50 respectively.

Statistical analysis Data between groups were compared using Mann-Whitney U tests. P-values less than 0.05 were considered statistically significant.Data were presented as box plots (Tukey) indicating the median and interquartile ranges (IQR). Whiskers represent 1.5 IQR, black dots represent outliers. Statistical analyses were performed with SPSS V21.0 (SPSS Inc, Chicago, USA).

46 SERUM INFLAMMATORY MARKERS NOT IN EARLY SpA RESULTS

The baseline characteristics of cohort 1 are shown in Table 1. Of note, 119 patients fulfilled the ASAS axSpA criteria (of which 36 fulfilled the imaging arm) and 191 did not fulfill the ASAS axSpA criteria. 3 CRP, ESR and calprotectin levels in early axSpA We first assessed whether CRP, ESR, and calprotectin levels, known to be elevated in full-blown ankylosing spondylitis, were also elevated in patients with early back pain fulfilling the ASAS axSpA criteria.42 The median (IQR) levels of CRP (3.00 [3.00-8.00] mg/l versus 3.00 [3.00-6.00] mg/l, p=0.317) (Figure 1A) and ESR (8.50 [5.00-17.00] mm/hour versus 9.00 [2.75-14.00] mm/hour, p=0.208) (Figure 1B) were not significantly different between patients fulfilling and not fulfilling the ASAS axSpA criteria in the SPACE cohort. In contrast, calprotectin levels (294 [214-367] ng/ml versus 251 [196-339] ng/ml, p=0.01) were significantly higher in patients with early back pain fulfilling versus those not fulfilling the ASAS axSpA criteria (Figure 1C). However, the discriminating value of calprotectin at the individual level was low: using a cut off value for calprotectin at a specificity of 90.0%

7 A B C

C R P E S R C a lp ro te c tin 1 6 0 1 0 0 8 0 0 1 5 0 *

8 0 r 8 0

u 6 0 0

7 0 l l o

/ 6 0 6 0 h m g 5 0 / / 4 0 0 m 4 0 4 0 g m

3 0 n 2 0 0 2 0 m 2 0 1 0 0 0 0 A S A S + A S A S - A S A S + A S A S - A S A S + A S A S -

D E F

C R P ESR Calprotectin 1 6 0 100 800 1 5 0 * 8 0 80 7 0 600 l

/ 6 0 60 g 5 0 400 m 4 0 40 3 0 ng/ml 2 0 200

mm/hour 20 1 0 0 0 0 A S A S + A S A S - ASAS+ imaging arm ASAS- ASAS+ imaging arm ASAS-

Figure 1. Serum levels of A, c-reactive protein (CRP), B, erythrocyte sedimentation rate (ESR), and C, calprotectin of cohort 1 (SPACE) with early back pain patients fulfilling the ASAS axial spondyloarthritis (axSpA) criteria (n=119) versus early back pain patients not fulfilling the ASAS axial SpA criteria (n=191); and serum levels of D, CRP, E, ESR, and F, calprotectin of cohort 1 (SPACE) with patients fulfilling the ASAS axSpA criteria according to the imaging arm (n=36) versus patients not fulfilling the ASAS axial SpA criteria (n=191).Boxplot (Tukey): Data are presented as median (interquartile range). Whiskers represent 1.5 IQR and black dots represent outliers. *p<0.05 by Mann-Whitney U test.

CHAPTER 3 47 (412.40 ng/ml), the sensitivity was 10.0% and the positive predictive value (PPV) 38.7%. The post-test probability of having axSpA is thus not increased in comparison with the pre-test probability (119 out of 310 or 38.4%).

CRP, ESR and calprotectin levels in the imaging arm of early axSpA The ASAS criteria consist of an imaging and a clinical arm. The imaging arm can be fulfilled only if abnormal MRI (according to the ASAS/OMERACT definition) or X-ray of sacroiliiac (SI-) joints (according to mNY) abnormalities are visible, and since calprotectin is an independent marker for axial spinal progression it might better associate to the imaging arm of the ASAS criteria. Therefore, we conducted similar analyses in patients fulfilling the imaging arm (n=36) of the ASAS axSpA criteria versus those not fulfilling the criteria (n=191). Again, levels of CRP (4.00 [3.00-9.00] mg/l versus 3.00 [3.00-6.00] mg/l, p=0.175) (Figure 1D), and ESR (6.50 [4.250-19.00] mm/hour versus 9.00 [4.00-13.00] mm/hour, p=0.512) (Figure 1E) were not different between the early axSpA patients fulfilling the ASAS imaging arm and controls not fulfilling the ASAS criteria (Figure 1D and E). Levels of calprotectin (313 [237- 371] ng/ml versus 253 [195-338] ng/ml, p=0.01) were significantly higher in early back pain patients fulfilling the imaging arm of the ASAS axSpA criteria versus those not fulfilling the criteria (Figure 1F). As for the global axSpA group, however, the discriminatory value at the individual level was low. Using again a cut off for calprotectin at the 90% specificity level (249.95 ng/ml) as example, the PPV was 80% but the sensitivity was only 7.7%.

IL-27, hBD-2 and LCN-2 in full-blown AS versus healthy controls As CRP, ESR, and calprotectin were not useful as diagnostic biomarkers for axSpA in the SPACE cohort, we explored the potential value of three additional potential biomarkers. To this purpose, we first assessed their serum levels in established active AS (cohort 2) versus controls. IL-27 was undetectable in all but one serum samples of AS and healthy controls (data not shown). Serum levels of hBD-2 (median and IQR: 2.61 [0.00-8.93] ng/ml versus 0.00 [0.00-26.65] ng/ml, p=0.572) (Figure 2A) and LCN-2 (55.47 [38.14-66.30] ng/ml versus 51.82 [46.59-68.57] ng/ml, p=0.562) (Figure 2B) were clearly detectable but were not elevated in AS versus healthy controls. As none of the 3 makers was elevated in AS, we did not proceed with testing the serum levels in the early back pain (SPACE) cohort.

DISCUSSION

We set up this study to assess whether inflammatory serum biomarkers can contribute to the diagnosis of axSpA in individuals presenting with early back pain. A first important finding is that serum CRP levels and ESR are not elevated in patients with early axSpA versus patients with back pain from different origins, despite that fact that elevated CRP is one

48 SERUM INFLAMMATORY MARKERS NOT IN EARLY SpA A B

hBD-2 LCN-2 150 150

100 100 ng/ml ng/ml 50 50 3

0 0 AS Healthy Controls AS Healthy Controls

Figure 2 Serum levels of A, human beta defensin-2 (hBD-2) and B, lipocalin-2 (LCN-2) in active full-blown . ankylosing spondylitis (AS, n=21) versus healthy controls (n=19). Data are presented as median (interquartile range). Boxplot (Tukey): Data are presented as median (interquartile range). Whiskers represent 1.5 IQR and black dots represent outliers. *P<0.05 by Mann-Whitney U test. *p<0.05 by Mann-Whitney U test.

of the features included in the ASAS axSpA criteria. Accordingly, specificity, sensitivity, and PPV analyses did not allow discrimination between patients and controls (data not shown). Importantly, however, the value of CRP and ESR were assessed here in an univariate analysis as the aim was to identify biomarkers that can easily1 and reproducibly be used to screen patients with early back pain to diagnose axSpA early. Albeit CRP and ESR do not appear to be useful tools for this purpose, this does not imply that these markers cannot be useful in individual patient when combined with other SpA features in a multivariate approach.

A second interesting finding is that, in contrast to CRP and ESR, calprotectin levels were significantly increased in early axSpA versus controls with back pain. This finding is consistent with a series of previous studies suggesting that calprotectin may slightly outperform CRP as marker of tissue inflammation in SpA,26–28,43 potentially because this protein is released during infiltration of myeloid cells in tissues and may thus more directly reflect some of the pathological processes in SpA. However, the difference in serum calprotectin levels detected at the group level between early axSpA and controls was not robust enough in terms of sensitivity and specificity to translate in a useful discriminative tool toidentify axSpA patients. These findings are consistent with studies on other serum inflammatory markers including IL-6,44,45 alpha-2-macroglobulin,46 MMP-3,47–50 and PTX-3,51 which have been explored already in established axSpA but were not robust enough to apply in a diagnostic setting in early axSpA.

The lack of increase in serum levels of inflammatory biomarkers in axSpA versus controls is likely related to two issues. First, these biomarkers do not reflect a SpA-specific process but rather inflammation in general and therefore lack specificity. Intercurrent infections or the presence of another chronic inflammatory disease may lead to elevated CRP, ESR, and/ or calprotectin in patients with mechanical back pain. One way to circumvent this issue would be to measure factors that are more specifically related to the immunopathology

CHAPTER 3 49 of axSpA, such as biomarkers reflecting the activation of the IL-23/IL-17 axis. Despite their value in psoriasis and IBD, however, the IL-17-driven antimicrobial peptides hBD-2 and LCN- 2 were not elevated in axSpA. This might be due to the fact that these peptides are mainly produced by epithelial cells, including keratinocytes and gut epithelial cells, and that IL-17 may act on different cell types in axSpA. Further investigations in the mechanistic aspects of SpA immunopathology remain warranted to identify novel potential biomarkers.

A second reason for the absence of elevated serum levels of inflammatory markers such as CRP and ESR may be the fact that inflammation is restricted to specific tissue compartments and does not extend to the systemic circulation and/or lymphoid organs in SpA. This concept is supported by the facts that a) in the present study a majority of early axSpA patients have CRP and ESR levels within the normal range despite active disease (n=171, 55.2%), b) even in the subset of patients with positive MRI imaging demonstrating active inflammation in the SI joints, CRP, ESR, and calprotectin levels were not further increased, and c) also in established SpA with active disease CRP levels are elevated in only one-third of the patients.20–23 Accordingly, we previously focused our research on the immunopathology of affected tissues, such as the synovial membrane, and found striking and reproducible alterations which, however, were not reflected in the peripheral blood compartment.52,53 If this concept is correct, the search for diagnostic markers should probably be focused on non-invasive measurements of tissue pathology (including MRI, PET-CT, and other types of molecular imaging) rather than on serum biomarkers.

50 SERUM INFLAMMATORY MARKERS NOT IN EARLY SpA REFERENCES

1. Rudwaleit M, Haibel H, Baraliakos X, et al. The early patients with axial spondyloarthritis in primary care. J disease stage in axial spondylarthritis: results from the Rheumatol 2011;38(11):2452–60. German Spondyloarthritis Inception Cohort. Arthritis 14. Van Hoeven L, Vergouwe Y, de Buck PDM, Luime JJ, Rheum 2009;60(3):717–27. Hazes JMW, Weel AE a M. External Validation of a 2. Wallis D, Haroon N, Ayearst R, Carty A, Inman RD. Referral Rule for Axial Spondyloarthritis in Primary 3 Ankylosing spondylitis and nonradiographic axial Care Patients with Chronic Low Back Pain. PLoS One spondyloarthritis: part of a common spectrum or 2015;10(7):e0131963. distinct diseases? J Rheumatol 2013;40(12):2038–41. 15. Rudwaleit M, Jurik a G, Hermann K-G a, et al. Defining 3. Boonen A, Sieper J, van der Heijde D, et al. The burden active sacroiliitis on magnetic resonance imaging (MRI) of non-radiographic axial spondyloarthritis. Semin for classification of axial spondyloarthritis: a consensual Arthritis Rheum 2015;44(5):556–62. approach by the ASAS/OMERACT MRI group. Ann 4. Van Den Bosch F, Kruithof E, Baeten D, et al. Randomized Rheum Dis 2009;68(10):1520–7. double-blind comparison of chimeric monoclonal 16. Rudwaleit M, Khan M a, Sieper J. The challenge antibody to tumor necrosis factor alpha (infliximab) of diagnosis and classification in early ankylosing versus placebo in active spondylarthropathy. Arthritis spondylitis: do we need new criteria? Arthritis Rheum Rheum 2002;46(3):755–65. 2005;52(4):1000–8. 5. Braun J, Brandt J, Listing J, et al. Treatment of 17. Feldtkeller E, Khan MA, van der Heijde D, van der active ankylosing spondylitis with infliximab: a Linden S, Braun J. Age at disease onset and diagnosis randomised controlled multicentre trial. Lancet delay in HLA-B27 negative vs. positive patients with 2002;359(9313):1187–93. ankylosing spondylitis. Rheumatol Int 2003;23(2):61–6. 6. Sieper J, van der Heijde D, Dougados M, et al. Efficacy 18. Sorensen J, Hetland ML. Decreases in diagnostic delay and safety of adalimumab in patients with non- are supported by sensitivity analyses. Ann Rheum Dis radiographic axial spondyloarthritis: results of a 2014;73(7):e45–e45. randomised placebo-controlled trial (ABILITY-1). Ann 19. Salvadorini, F. Bandinelli, A. Delle Sedie, L. Riente, A. Rheum Dis 2013;72(6):815–22. Candelieri, S. Generini, N. Possemato, S. Bombardieri 7. Van der Heijde D, Landewé R, Einstein S, et al. MM-C. Ankylosing spondylitis: how diagnostic and Radiographic progression of ankylosing spondylitis therapeutic delay have changed over the last six after up to two years of treatment with etanercept. decades. Clin Exp Rheumatol 2012;30(4):561–5. Arthritis Rheum 2008;58(5):1324–31. 20. Benhamou M, Gossec L, Dougados M. Clinical relevance 8. Van der Heijde D, Salonen D, Weissman BN, et al. of C-reactive protein in ankylosing spondylitis and Assessment of radiographic progression in the spines evaluation of the NSAIDs/coxibs’ treatment effect of patients with ankylosing spondylitis treated with on C-reactive protein. Rheumatology (Oxford) adalimumab for up to 2 years. Arthritis Res Ther 2010;49(3):536–41. 2009;11(4):R127. 21. Visvanathan S, Wagner C, Marini JC, et al. Inflammatory 9. Van der Heijde D, Landewé R, Baraliakos X, et al. biomarkers, disease activity and spinal disease Radiographic findings following two years of infliximab measures in patients with ankylosing spondylitis therapy in patients with ankylosing spondylitis.Arthritis after treatment with infliximab. Ann Rheum Dis Rheum 2008;58(10):3063–70. 2008;67(4):511–7. 10. Maas F, Spoorenberg A, Brouwer E, et al. Spinal 22. De Vries MK, van Eijk IC, van der Horst-Bruinsma radiographic progression in patients with ankylosing IE, et al. Erythrocyte sedimentation rate, C-reactive spondylitis treated with TNF-α blocking therapy: a protein level, and serum amyloid a protein for patient prospective longitudinal observational cohort study. selection and monitoring of anti-tumor necrosis factor PLoS One 2015;10(4):e0122693. treatment in ankylosing spondylitis. Arthritis Rheum 11. Maksymowych WP, Zheng Y, Wichuk S, 2009;61(11):1484–90. Chiowchanwisawakit P, Lambert RG. The effect of TNF 23. Poddubnyy D a, Rudwaleit M, Listing J, Braun J, Sieper inhibition on radiographic progression in ankylosing J. Comparison of a high sensitivity and standard apondylitis: An observational cohort study of 384 C reactive protein measurement in patients with patients. Ann Rheum Dis 2015;74(Suppl2): 123. ankylosing spondylitis and non-radiographic axial 12. Brandt HC, Spiller I, Song I-H, Vahldiek JL, Rudwaleit spondyloarthritis. Ann Rheum Dis 2010;69(7):1338–41. M, Sieper J. Performance of referral recommendations 24. Kane D, Roth J, Frosch M, Vogl T, Bresnihan B, FitzGerald in patients with chronic back pain and suspected axial O. Increased perivascular synovial membrane spondyloarthritis. Ann Rheum Dis 2007;66(11):1479– expression of myeloid-related proteins in psoriatic 84. arthritis. Arthritis Rheum 2003;48(6):1676–85. 13. Poddubnyy D, Vahldiek J, Spiller I, et al. Evaluation 25. Kruithof E, De Rycke L, Vandooren B, et al. of 2 screening strategies for early identification of Identification of synovial biomarkers of response to

CHAPTER 3 51 experimental treatment in early-phase clinical trials an observational study. Ann Rheum Dis 2015;207544. in spondylarthritis. Arthritis Rheum 2006;54(6):1795– 39. Dougados M, Baeten D. Spondyloarthritis. Lancet 804. 2011;377(9783):2127–37. 26. De Rycke L, Baeten D, Foell D, et al. Differential 40. Van den Berg R, de Hooge M, van Gaalen F, Reijnierse expression and response to anti-TNFalpha treatment of M, Huizinga T, van der Heijde D. Percentage of patients infiltrating versus resident tissue macrophage subsets with spondyloarthritis in patients referred because of in autoimmune arthritis. J Pathol 2005;206(1):17–27. chronic back pain and performance of classification 27. Turina MC, Yeremenko N, Paramarta JE, De Rycke L, criteria: experience from the Spondyloarthritis Baeten D. Calprotectin (S100A8/9) as serum biomarker Caught Early (SPACE) cohort. Rheumatology (Oxford) for clinical response in proof-of-concept trials in axial 2013;52(8):1492–9. and peripheral spondyloarthritis. Arthritis Res Ther 41. Salinas GF, De Rycke L, Barendregt B, et al. Anti-TNF 2014;16(5):413. treatment blocks the induction of T cell-dependent 28. Turina MC, Sieper J, Yeremenko N, et al. Calprotectin humoral responses. Ann Rheum Dis 2013;72(6):1037– serum level is an independent marker for radiographic 43. spinal progression in axial spondyloarthritis. Ann 42. Sieper J, Rudwaleit M, Baraliakos X, et al. The Rheum Dis 2014;73(9):1746–8. Assessment of SpondyloArthritis international Society 29. Lin T-T, Lu J, Qi C-Y, et al. Elevated serum level of IL- (ASAS) handbook: a guide to assess spondyloarthritis. 27 and VEGF in patients with ankylosing spondylitis Ann Rheum Dis 2009;68 Suppl 2:ii1–44. and associate with disease activity. Clin Exp Med 43. Kruithof E, De Rycke L, Roth J, et al. Immunomodulatory 2014;15(2):227-231. effects of etanercept on peripheral joint synovitis 30. Gambichler T, Bechara FG, Scola N, Rotterdam S, in the spondylarthropathies. Arthritis Rheum Altmeyer P, Skrygan M. Serum levels of antimicrobial 2005;52(12):3898–909. peptides and proteins do not correlate with psoriasis 44. Gratacós J, Collado A, Filella F, et al. Serum cytokines (IL- severity and are increased after treatment with fumaric 6, TNF-alpha, IL-1 beta and IFN-gamma) in ankylosing acid esters. Arch Dermatol Res 2012;304(6):471–4. spondylitis: a close correlation between serum IL-6 31. Jansen P a M, Rodijk-Olthuis D, Hollox EJ, et al. and disease activity and severity. Br J Rheumatol Beta-defensin-2 protein is a serum biomarker for 1994;33(10):927–31. disease activity in psoriasis and reaches biologically 45. Pedersen SJ, Hetland ML, Sørensen IJ, Ostergaard relevant concentrations in lesional skin. PLoS One M, Nielsen HJ, Johansen JS. Circulating levels of 2009;4(3):e4725. interleukin-6, vascular endothelial growth factor, 32. Kamata M, Tada Y, Tatsuta a, et al. Serum lipocalin-2 YKL-40, matrix metalloproteinase-3, and total levels are increased in patients with psoriasis. Clin Exp aggrecan in spondyloarthritis patients during 3 years Dermatol 2012;37(3):296–9. of treatment with TNFα inhibitors. Clin Rheumatol 33. Romaní J, Caixàs A, Ceperuelo-Mallafré V, et al. 2010;29(11):1301–9. Circulating levels of lipocalin-2 and retinol-binding 46. Surrall K, Bird H, Dixon J. Caeruloplasmin , prealbumin protein-4 are increased in psoriatic patients and and u2-macroglobulin as potential indices of disease correlated with baseline PASI. Arch Dermatol Res activity in different arthritides. Clin Rheumatol 2013;305(2):105–12. 1987;(3):64–9. 34. Oikonomou KA, Kapsoritakis AN, Theodoridou C, et 47. Yang C, Gu J, Rihl M, et al. Serum levels of matrix al. Neutrophil gelatinase-associated lipocalin (NGAL) metalloproteinase 3 and macrophage colony- in inflammatory bowel disease: association with stimulating factor 1 correlate with disease activity in pathophysiology of inflammation, established markers, ankylosing spondylitis.Arthritis Rheum2004;51(5):691– and disease activity. J Gastroenterol 2012;47(5):519– 9. 30. 48. Maksymowych WP, Landewé R, Conner-Spady B, et al. 35. Yeşil A, Gönen C, Senateş E, et al. Relationship between Serum matrix metalloproteinase 3 is an independent neutrophil gelatinase-associated lipocalin (NGAL) levels predictor of structural damage progression in and inflammatory bowel disease type and activity. Dig patients with ankylosing spondylitis. Arthritis Rheum Dis Sci 2013;58(9):2587–93. 2007;56(6):1846–53. 36. Hreggvidsdottir HS, Noordenbos T, Baeten DL. 49. Vandooren B, Kruithof E, Yu DTY, et al. Involvement Inflammatory pathways in spondyloarthritis. Mol of matrix metalloproteinases and their inhibitors in Immunol 2014;57(1):28–37. peripheral synovitis and down-regulation by tumor 37. Baeten D, Baraliakos X, Braun J, et al. Anti-interleukin- necrosis factor alpha blockade in spondylarthropathy. 17A monoclonal antibody secukinumab in treatment Arthritis Rheum 2004;50(9):2942–53. of ankylosing spondylitis: a randomised, double-blind, 50. Chen C-H, Lin K-C, Yu DTY, et al. Serum matrix placebo-controlled trial. Lancet 2013;382(9906):1705– metalloproteinases and tissue inhibitors of 13. metalloproteinases in ankylosing spondylitis: 38. Baraliakos X, Borah B, Braun J, et al. Long-term effects MMP-3 is a reproducibly sensitive and specific of secukinumab on MRI findings in relation to clinical biomarker of disease activity. Rheumatology (Oxford) efficacy in subjects with active ankylosing spondylitis: 2006;45(4):414–20.

52 SERUM INFLAMMATORY MARKERS NOT IN EARLY SpA 51. Deban L, Jaillon S, Garlanda C, Bottazzi B, Mantovani A. Pentraxins in innate immunity: lessons from PTX3. Cell Tissue Res 2011;343(1):237–49. 52. Noordenbos T, Yeremenko N, Gofita I, et al. Interleukin- 17-positive mast cells contribute to synovial inflammation in spondylarthritis. Arthritis Rheum 2012;64(1):99–109. 53. Yeremenko N, Noordenbos T, Cantaert T, et al. Disease- specific and inflammation-independent stromal 3 alterations in spondylarthritis synovitis.Arthritis Rheum 2013;65(1):174–85.

CHAPTER 3 53

Clinical and imaging signs of spondyloarthritis in first-degree relatives of HLA-B27 positive ankylosing spondylitis patients: the pre- spondyloarthritis (Pre-SpA) cohort

Maureen C. Turina,1*, Janna J. de Winter 1*, Jaqueline E. Paramarta1, Mihaela Gamala1, Nataliya Yeremenko1,2, Marita N. Nabibux3, Robert Landewé1, Dominique L. Baeten1,2

1Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and immunology Center, Academic Medical Center/University of Amsterdam, 4 Amsterdam, The Netherlands; 2Department of Experimental Immunology, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; 3Department of Rheumatology, Tergooi Hospital, Hilversum, The Netherlands

*Equal contributions

Submitted for publication ABSTRACT

Objective To investigate if seemingly healthy first-degree relatives (FDRs) of ankylosing spondylitis (AS) patients have clinical, laboratory or imaging features of spondyloarthritis (SpA).

Methods FDRs of HLA-B27 positive AS patients between 18-40 years were included in Pre-SpA, a prospective inception cohort study. Clinical, biological and imaging data were obtained. FDRs were classified according to several sets of SpA classification criteria.

Results We report baseline features of 51 FDRs included in this study. Twenty-nine (57%) FDRs had back pain; 2 (4%) had psoriasis, and none had other extra-articular manifestations. Three (6%) FDRs had low grade sacroiliitis, 1 (2%) had cervical syndesmophytes on X-ray, and 11 (22%) had bone marrow edema on MRI of the sacroiliiac joints. Seventeen of 51 (33%) FDRs fulfilled SpA classification criteria: 7 (14%) both the ASAS axSpA and ESSG criteria, 6 (12%) only the ASAS axSpA, and 4 (8%) only the ESSG criteria. None fulfilled other SpA classification criteria. FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria had more frequently (inflammatory) back pain and had a higher level of disease activity. No difference in inflammatory parameters, peripheral and extra-articular disease and HLA-B27 presence was found. Six (16%) of the FDRs not fulfilling the ASAS axSpA and/or ESSG classification criteria had imaging abnormalities suggestive of SpA.

Conclusion A substantial proportion of seemingly healthy FDRs of HLA-B27 positive AS patients have clinical and/or imaging abnormalities suggestive of SpA. Thirty-three percent could be classified as SpA. Further follow-up will show which FDRs will develop clinically manifest SpA.

56 Pre-SpA INCEPTION COHORT INTRODUCTION

A major feature of spondyloarthritis (SpA) is its insidious onset and slow progression of signs and symptoms as well as of structural damage.1 This leads to a significant diagnostic delay of five to ten years.2-4 More recently, the development of referral strategies for early inflammatory back pain (IBP)5-7 and the use of magnetic resonance imaging (MRI) to visualize active sacroiliitis have probably decreased this diagnostic delay. However, detecting the earliest phases of the disease remains challenging, which has three potentially important 4 implications. First, the absence of an early diagnosis delays adequate treatment of signs and symptoms of inflammation for several months to years. Second, uncontrolled early disease may initiate osteoproliferation, a process seemingly resistant to treatment in the later phases.8-10 And third, when the earliest phase of the disease is ignored, a comprehensive chronological and hierarchical mapping of all cellular and molecular mechanisms that drive SpA is illusive.11,12

Similar to rheumatoid arthritis, SpA might have a subclinical phase of disease preceding the clinical established phase, where abnormalities are present but clinical symptoms are lacking. A systematic screening and meticulous follow-up of those that are at high risk of developing SpA may help to address such challenges. Models explaining the genetic susceptibility to ankylosing spondylitis (AS), the prototypical axial form of SpA, suggest an 8% recurrence risk in first-degree relatives (FDRs) of AS patients.13 The comparison of familial and sporadic cases of AS suggest a higher familial aggregation of AS in patients with HLA-B27 positivity.14 Accordingly, several studies have suggested a 10-12% recurrence risk of AS (as defined by the presence of radiographic sacroiliitis) in FDRs of AS patients, with a twofold higher risk in HLA-B27 positive individuals.15-18 Other SpA subtypes were also frequently observed in FDRs of probands with AS or SpA.18,19

The fact that all these cross-sectional studies consistently show that FDRs of HLA-B27 positive AS patients have a strongly increased risk of developing SpA, opens the venue for systematic and detailed characterization of the earliest phases of SpA by prospective analysis of such FDRs at risk. In order to better detect, understand, and eventually treat the earliest phases of disease, the purpose of the present inception cohort study is to prospectively study the appearance and development of clinical, biological and imaging (X-ray and MRI) features of SpA in seemingly healthy FDRs of AS patients. We report here the baseline analysis of clinical and imaging features in 51 FDRs included in the study.

CHAPTER 4 57 PATIENTS AND METHODS

Study design Pre-SpA is an ongoing, prospective five-year inception cohort study. FDRs of HLA-B27 positive AS patients were included, between the ages of 18 and 40 years since new onset of SpA after the age of 45 years is rare and since the major objective is to study the development rather than the incidence of SpA. We selected probands with established AS rather than SpA as a whole spectrum for 3 reasons. 1) We aimed for a high entry-threshold by only including HLA-B27 positive AS patients, thereby avoiding as much misdiagnosed probands as possible, 2) the other genetic risk factors besides HLA-B27 have largely only been established in AS and not in other SpA subtypes,20 and 3) HLA-B27 positive patients tend to have more ankylosis than HLA-B27 negative patients,21 thereby possibly favoring the presence and detection of structural radiographic changes in FDRs. All participants signed informed consent prior to any study procedures. Major exclusion criteria were the presence of already diagnosed SpA, other rheumatic conditions including fibromyalgia, and back pain due to other conditions, such as intervertebral disc degeneration. All study procedures were done in compliance with the Helsinki Declaration. The study protocol was approved by the Medical Ethics Committee of the Academic Medical Center/University of Amsterdam.

Clinical evaluation at baseline Data on demographics (gender, age, and race), medical and family history, and the use of medication were recorded. History specific for SpA-related features included back pain, IBP as defined by the Assessment of SpondyloArthritis international Society (ASAS) criteria,22 peripheral arthritis, enthesitis, dactylitis, psoriasis, inflammatory bowel disease (IBD), and uveitis, all objectified by a physician. Disease activity was measured by physician’s global assessment of disease activity (PhGA), patient’s global assessment of disease activity (PGA) on a visual analogue score (VAS), by patient’s nocturnal pain VAS, Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and by Ankylosing Spondylitis Disease Activity Score-CRP (ASDAS-CRP). Function was assessed by the Bath Ankylosing Spondylitis Functional Index (BASFI). Clinical examination included a 68-tender and a 66-swollen joint count (TJC and SJC), an enthesitis score (Maastricht Ankylosing Spondylitis Enthesitis Score (MASES)), a dactylitis evaluation, the linear Bath Ankylosing Spondylitis Metrology Index (BASMI), chest expansion, and occiput-to-wall distance.22 Cut-off values of modified Schober and chest expansion were set at 4.5 cm and 3.6 cm respectively.23

Biological evaluation at baseline Laboratory assessments included HLA-B27 testing, C-reactive protein (CRP) serum levels, and erythrocyte sedimentation rate (ESR). Levels of calprotectin were determined by enzyme-

58 Pre-SpA INCEPTION COHORT linked immunosorbent assay (ELISA) as indicated by the manufacturer (Hycult Biotech, the Netherlands).24,25

Imaging evaluation at baseline Radiographs of the lumbar and cervical spine and of the sacroiliac joints (SIJs) were obtained. Anteroposterior X-rays of the SIJs were scored according to the modified New York (mNY) criteria.26 Lateral and anteroposterior X-rays of the lumbar and cervical spine were scored according to the modified Stoke AS Spine Score (mSASSS).27 MRIs were performed 4 on a 3.0 tesla scanner (Philips Medical Systems, Best, The Netherlands), with sagittal T-1 weighted and short tau inversion recovery (STIR) sequences with a slice thickness of 4 mm for the spine (sagittal) and the SIJs (semicoronal). Both X-rays and MRIs were scored by one experienced, blinded reader (RL) for the presence of bone marrow edema (BME), according to the ASAS/OMERACT definition of a positive MRI.28

Analysis Clinical, biological and imaging features of SpA were reported descriptively. Apart from missing imaging data of one FDR, data were complete. FDRs were classified according to the ASAS classification criteria for axial SpA (axSpA),29 the ASAS peripheral SpA criteria,30 the European Spondyloarthropathy Study Group (ESSG) criteria,31 the Amor criteria,32 the ClASsification criteria for Psoriatic ARthritis 33(CASPAR), and the mNY criteria.26 The proportion of FDRs fulfilling any SpA classification criteria was compared to that of FDRs not fulfilling SpA classification criteria by using the chi-square test. For continuous data Mann Whitney-U test was used. Data were presented as numbers and percentages or as mean and standard deviation (SD). Statistical tests were 2-sided and P-values less than 0.05 were considered statistically significant. A similar analysis was performed for FDRs with imaging signs suggestive of SpA versus FDRs without imaging abnormalities.

RESULTS

Clinical and biological features of SpA in FDRs of HLA-B27 positive AS patients Fifty-one FDRs from 38 probands (comprising 33 parents and 5 siblings) with ASwere included in this analysis. Main reasons to refuse study participation were abroad residencies, being too busy, fear for blood collection, and a few FDRs were not interested to participate without clarifying their motivation. The mean (SD) age at inclusion was 25 (5) years; 25 (49%) participants were male, and 26 (51%) were HLA-B27 positive. The baseline characteristics of the study population are summarized in Table 1. Twenty-nine (57%) FDRs reported back pain after inquiry, and 11 (22%) fulfilled the IBP criteria. Of the 51 FDRs, 20 (40%) reported

CHAPTER 4 59 Table 1. Clinical characteristics, laboratory and imaging results of the FDRs of those fulfilling and not fulfilling the ASAS axSpA and/or ESSG classification criteria Fulfilling vs not fulfilling Study Fulfilling the ASAS axSpA and/or the ASAS axSpA and/or population ESSG criteria ESSG criteria

(n= 51) Yes (n=17) No (n=34) p-value Demographics Male gender 25 (49) 7 (41) 18 (53) 0.433 Caucasian 50 (98) 16 (94) 34 (100) 0.157 Age, years, mean (SD) 25.2 (5.1) 25.8 (5.2) 24.9 (5) 0.595 2 BMI, kg/m , mean (SD) 23.53 (3.16) 23.57 (3.72) 23.51 (2.89) 0.935 HLA-B27 positive 26 (51) 8 (47) 18 (53) 0.695 Current smoker 14 (28) 4 (24) 10 (29) 0.973 Use of DMARD, TNF inhibitor 0 (0) 0 (0) 0 (0) - or corticosteroids ever

History Axial disease Back pain 29 (56.9) 17 (100) 12 (35) <0.001 Inflammatory back pain 11 (21.6) 11 (65) 0 (0) <0.001 Buttock pain 2 (3.9) 1 (6) 1 (3) 0.614 Good response to NSAIDs 5 (9.8) 5 (29) 0 (0) 0.001 Peripheral disease Arthralgia (past/present) 20 (40) 9 (53) 11 (32) 0.160 Peripheral arthritis (past/ present) 1 (2) 0 (0) 1 (3) 0.480 Enthesitis(past/present) 0 (0) 0 (0) 0 (0) - Dactylitis (past/present) 0 (0) 0 (0) 0 (0) - Extra-articular disease Psoriasis (past/present) 2 (4) 2 (12) 0 (0) 0.043 IBD (past/present) 1 (2) 1 (6) 0 (0) 0.157 Urethritis/diarrhea (past/ 0 (0) 0 (0) 0 (0) present) - Uveitis (past/present) 1 (2) 1 (6) 0 (0) 0.157

Family history Ankylosing spondylitis 51 (100) 17 (100) 34 (100) - Psoriatic arthritis 2 (4) 2 (12) 0 (0) 0.043 Psoriasis 2 (4) 0 (0) 2 (6) 0.312 Inflammatory bowel disease 6 (12) 3 (18) 3 (9) 0.361 Uveitis 1 (2) 0 (0) 1 (3) 0.480

Disease activity measurements PhGA, 0-100mm VAS, mean (SD) 7 (11) 12 (3) 4 (9) <0.001 PGA, 0-100mmVAS, mean (SD) 11 (16) 15 (18) 8 (15) 0.065 Patient nocturnal pain, 5 (14) 13 (22) 2 (3) 0.003 0-100mm VAS, mean (SD)

60 Pre-SpA INCEPTION COHORT BASDAI, 0-10cm, mean (SD) 1.17 (1.30) 2.04 (1.42) 0.73 (1.00) <0.001 BASDAI >4 3 (6) 2 (12) 1 (3) 0.211 ASDAS-CRP, mean (SD) 0.41 (0.49) 0.67 (0.59) 0.27 (0.38) 0.007 ASDAS-CRP ≥1.3 3 (6) 2 (12) 1 (3) 0.211 BASFI, 0-10cm, mean (SD) 0.43 (0.77) 0.85 (1.08) 0.22 (0.46) 0.001

Clinical examination modified Schober, cm, mean 6.43 (6.21) 5.79 (4.03) 6.62 (7.09) 0.571 (SD) modified Schober <4.5cm 11 (22) 5 (29) 6 (18) 0.336 4 Chest expansion, cm, mean (SD) 6.26 (2.48) 5.97 (2.25) 6.04 (2.61) 0.628 Chest expansion <3.6cm 2 (4) 1 (6) 1 (3) 0.614 BASMI, 0-10, mean (SD) 2.16 (0.76) 2.35 (0.93) 2.06 (0.65) 0.098 TJC > 0, 0-68 joints 8 (16) 5 (29) 3 (9) 0.059 SJC > 0, 0-66 joints 0 (0) 0 (0) 0 (0) - MASES>0 10 (20) 7 (41) 3 (9) 0.007 Dactylitis, presence 0 (0) 0 (0) 0 (0) - Nail psoriasis, presence 0 (0) 0 (0) 0 (0) - Psoriasis, presence 0 (0) 0 (0) 0 (0) -

Laboratory CRP, mg/L, mean (SD) 2.37 (3.45) 1.94 (2.71) 2.58 (3.79) 0.704 CRP above 5 mg/L 6 (12) 1 (6) 5 (15) 0.361 ESR, mm/h, mean (SD) 6.72 (7.16) 6.41 (6.16) 6.88 (7.71) 0.874 ESR above 20 mm/h 4 (8) 1 (6) 3 (9) 0.695 345.44 Calprotectin, ng/mL, mean (SD) 310.03 (136.07) 363.69 (124.71) 0.169 (129.85) Calprotectin above 715 ng/ml 0 (0) 0 (0) 0 (0) -

Imaging Low grade sacroiliitis on X-ray 3 (6) 3 (18) 0 (0) 0.014 Syndesmophytes of the lumbar 0 (0) 0 (0) 0 (0) spine on X-ray - Syndesmophytes of the cervical 1 (2) 0 (0) 1 (3) 0.473 spine on X-ray Inflammatory lesions of the SIJ 11 (22) 7 (41) 4 (12) 0.020 on MRI Inflammatory lesions of the 0 (0) 0 (0) 0 (0) spine on MRI - ASAS, Assessment of SpondyloArthritis international Society; ASDAS, Ankylosing Spondylitis Disease Activity Score; AxSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; BMI, body mass index; cm, centimeter; CRP, c-reactive protein; DMARD, disease modifying antirheumatic drugs; ESR, erythrocyte sedimentation rate; ESSG, European Spondyloarthropathy Study Group; IBD, inflammatory bowel disease; HLA-B27, human leukocyte antigen-B27; kg, kilogram; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; m, meter; mm, millimeter; MRI, magnetic resonance imaging; NSAIDs, non-steroidal anti-inflammatory drugs; PGA patient global assessment of disease activity; PhGA, physician global assessment of disease activity; SD, standard deviation; SIJ, sacroiliac joints; SJC, swollen joint count; SpA, spondyloarthritis; TJC, tender joint count; TNF, tumor necrosis factor; VAS, visual analogue score. Data presented as n (%) unless otherwise specified.

CHAPTER 4 61 past or present arthralgia, 1 (2%) reported past or present peripheral arthritis. None of the FDRs reported past or present enthesitis or dactylitis. Of all 51 FDRs, 2 (4%) had psoriasis, 1 (2%) reported IBD, no FDR reported past or present urethritis/diarrhea, and 1 (2%) had past or present uveitis. A modified Schober of <4.5 cm was found in 11 (22%) FDRs, and 2 (4%) FDRs had a chest expansion of <3.6 cm. None of the FDRs had arthritis but 8 (16%) had at least one tender joint. Ten (20%) participants had tenderness of at least one entheseal point, but none had dactylitis, nail psoriasis, or skin psoriasis. Laboratory results showed an elevated CRP, ESR or calprotectin in 6 (12%), 4 (8%) and 0 (0%) of the FDRs, respectively.

Imaging features of SpA in FDRs of HLA-B27 positive AS patients Three (6%) FDRs had low-grade sacroiliitis not fulfilling the mNY criteria (grade I uni- or bilaterally, or grade II unilaterally) on X-ray, and 1 (2%) FDR had syndesmophytes on spinal X-ray. MRI showed inflammatory lesions of the SIJ, and spine in 11 (22%), and 0 (0%) of the FDRs, respectively (Figure 1).

Application of SpA classification criteria to FDRs of HLA-B27 positive AS patients Seventeen (33%) FDRs fulfilled any of the SpA classification criteria at baseline: 7(14%) fulfilled both the ASAS axSpA and the ESSG criteria, 6 (12%) only fulfilled the ASAS axSpA, and 4 (8%) only fulfilled the ESSG criteria. Of these 17 FDRs, 3 (6%) FDRs also fulfilled the Amor criteria. None of the FDRs fulfilled the ASAS peripheral SpA criteria, the CASPAR, or the mNY criteria.

When comparing FDRs fulfilling with those not fulfilling the ASAS axSpA and/or ESSG classification criteria (Table 1), back pain (17 (100%) vs. 12 (35%), p<0.001) and IBP(11 (65%) vs. 0 (0%), p<0.001) were more prevalent in the former group. Good response to non- steroidal anti-inflammatory drugs (NSAIDs) was also more prevalent in the group of FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria compared to those who did not (5 (29%) vs. 0 (0%), p=0.001). Furthermore, FDRs who fulfilled the ASAS axSpA and/or ESSG classification criteria had a higher number of tender entheseal points, and showed a trend to have more tender joints when compared with those not fulfilling the ASAS axSpA and/ or ESSG classification criteria (7 (41%) vs. 3 (9%), p=0.007 and 5 (29%) vs. 3 (9%), p=0.059, respectively). In contrast, there were no differences in peripheral and extra-articular disease (past or present) between both groups. The group fulfilling the ASAS axSpA and/or ESSG classification criteria had a higher mean (SD) PhGA (12 (3) vs. 4 (9) mm, p<0.001), a higher VAS for patient’s nocturnal pain (13 (22) vs. 2 (3) mm; p=0.003), a higher BASDAI (2.04 (1.42) vs. 0.73 (1.00), p<0.001), and a higher ASDAS-CRP (0.67 (0.59) vs. 0.27 (0.38), p=0.007), with a trend towards a higher PGA (15 (18) vs. 8 (15) mm, p=0.065). The BASFI was worse in the FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria (0.85 (1.08) vs. 0.22 (0.46), p=0.001). FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria had low overall disease activity scores: only three FDRs had a BASDAI >4 and/or ASDAS-CRP >1.3.

62 Pre-SpA INCEPTION COHORT A

4

B

Figure 1. Images of cervical syndesmophytes on lateral X-ray (A) and a representative image of bone marrow edema on the magnetic resonance imaging of the sacroiliac joints (MRI SIJ) in the STIR sequences (B).

Inflammatory serum markers were not increased in FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria. However, BME on MRI (7 (41%) vs. 4 (12%), p=0.020) and signs of sacroiliitis on X-ray (3 (18%) vs. 0 (0%), p=0.014) were found more often in those fulfilling the ASAS axSpA and/or ESSG classification criteria.

Concluding, FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria had more axial, entheseal, and joint pain which was reflected in higher disease activity and worse function than those not fulfilling the ASAS axSpA and/or ESSG classification criteria. Except for BME

CHAPTER 4 63 on MRI, however, they did not show objective signs of inflammation such as clinical arthritis or elevated acute phase response.

Comparison of FDRs with and without imaging abnormalities Interestingly, 6 of 38 (16%) FDRs not fulfilling the ASAS axSpA and/or ESSG classification criteria had imaging abnormalities suggestive of SpA: One FDR had syndesmophytes on the cervical spine X-ray, and 5 FDRs had BME on MRI of the SIJ (Table 1 and Figure 1). Therefore, we conducted an analysis of all FDRs with imaging abnormalities (the group fulfilling the ASAS axSpA and/or ESSG classification criteria and the group without symptoms but with imaging abnormalities) versus those without imaging abnormalities. However, there were no differences in demographics, history of SpA symptoms, family history, disease activity, clinical examination, and laboratory findings between these two groups (Table 2).

Comparison of HLA-B27 positive versus HLA-B27 negative FDRs Previous studies have shown that SpA mainly manifests in HLA-B27 positive FDRs.13,18 Therefore, we performed another analysis comparing the HLA-B27 positive FDRs with the HLA-B27 negative FDRs. Nevertheless, there were no differences found between these groups in demographics, history of SpA symptoms, disease activity, clinical examination, laboratory and imaging data (Table 3).

64 Pre-SpA INCEPTION COHORT Table 2. Clinical characteristics, laboratory and imaging results of the FDRs with and without imaging abnormalities on X-ray and/or MRI of the SIJ and spine Imaging abnormalities

Yes (n= 12) No (n= 38) p-value Demographics Male gender 6 (50) 19 (50) - Caucasian 12 (100) 37 (98) 0.574 Age, years, mean (SD) 26.75 (5.88) 25.45 (5.06) 0.480 BMI, kg/m2, mean (SD) 23.92 (3.91) 23.31 (2.90) 0.667 HLA-B27 positive 6 (50) 19 (50) - 4 Current smoker 3 (25) 11 (29) 0.910 Use of DMARD, TNF inhibitors or 0 (0) 0 (0) - corticosteriods ever

History Axial disease Back pain 8 (67) 21 (65) 0.490 Inflammatory back pain 3 (25) 8 (21) 0.776 Buttock pain 0 (0) 2 (5) 0.416 Good response to NSAIDs 3 (25) 2 (5) 0.049

Peripheral disease Arthralgia (past/present) 5 (42) 15 (40) 0.894 Peripheral arthritis (past/present) 0 (0) 1 (3) 0.574 Enthesitis(past/present) 0 (0) 0 (0) - Dactylitis (past/present) 0 (0) 0 (0) - Extra-articular disease Psoriasis (past/present) 0 (0) 2 (5) 0.422 IBD (past/present) 1 (8) 0 (0) 0.075 Urethritis/diarrhoea (past/present) 0 (0) 0 (0) - Uveitis (past/present) 0 (0) 1 (3) 0.574

Family history Ankylosing spondylitis 12 (100) 38 (100) - Psoriatic arthritis 0 (0) 2 (5) 0.422 Psoriasis 0 (0) 2 (5) 0.422 Inflammatory bowel disease 3 (25) 3 (8) 0.116 Uveitis 0 (0) 1 (3) 0.574

Disease activity measurements PhGA, 0-100mm VAS, mean (SD) 10 (14) 6 (11) 0.587 PGA, 0-100mm VAS, mean (SD) 15 (19) 9 (16) 0.231 Patient nocturnal pain, 0-100mm 9 (25) 4 (8) 0.530 VAS, mean (SD) BASDAI, 0-10cm, mean (SD) 1.48 (1.68) 1.18 (1.18) 0.593 BASDAI >4 1 (8) 2 (5) 0.699 ASDAS-CRP, mean (SD) 0.55 (0.67) 0.36 (0.43) 0.376 ASDAS-CRP ≥1.3 2 (16) 1 (3) 0.077

CHAPTER 4 65 BASFI, 0-10cm, mean (SD) 0.61 (1.27) 0.39 (0.56) 0.563

Clinical examination Modified Schober, cm, mean (SD) 6.38 (4.64) 6.39 (6.75) 0.716 Modified Schober <4.5cm 2 (17) 8 (21) 0.743 Chest expansion, cm, mean (SD) 7.50 (2.82) 5.93 (2.28) 0.078 Chest expansion <3.6cm 1 (8) 1 (3) 0.384 BASMI, 0-10, mean (SD) 2.25 (0.45) 2.11 (0.83) 0.714 TJC > 0, 0-68 joints 1 (8) 7 (18) 0.411 SJC > 0, 0-66 joints 0 (0) 0 (0) - MASES>0 3 (25) 7 (18) 0.623 Dactylitis, presence 0 (0) 0 (0) - Nail psoriasis, presence 0 (0) 0 (0) - Psoriasis, presence 0 (0) 0 (0) -

Laboratory measurements CRP, mg/L, mean (SD) 3.19 (4.89) 2.10 (2.95) 0.820 CRP above 5 mg/L 2 (16) 4 (11) 0.572 ESR, mm/h, mean (SD) 7.00 (9.72) 6.65 (6.39) 0.639 ESR above 20 mm/h 3 (25) 1 (3) 0.981 Calprotectin, ng/mL, mean (SD) 383.31 (144.57) 337.27 (126.03) 0.307 Calprotectin above 715 ng/ml 0 (0) 0 (0) -

Classification criteria ASAS axSpA 6 (16) 7 (58) 0.004 ASAS peripheral SpA 0 (0) 0 (0) - ESSG 8 (22) 3 (25) 0.776 Amor criteria 1 (8) 2 (5) 0.699 CASPAR 0 (0) 0 (0) - mNY 0 (0) 0 (0) - ASAS, Assessment of SpondyloArthritis international Society; ASDAS, Ankylosing Spondylitis Disease Activity Score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; BMI, body mass index; CASPAR, ClASsification criteria for Psoriatic ARthritis; cm, centimeter; CRP, c-reactive protein; DMARD, disease modifying antirheumatic drugs; ESSG, European Spondyloarthropathy Study Group; ESR, erythrocyte sedimentation rate; IBD, inflammatory bowel disease; HLA-B27, human leukocyte antigen-B27; kg, kilogram; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; m, meter; mm, millimeter; mNY, modified New York; NSAIDs, non-steroidal anti-inflammatory drugs; PGA patient global assessment of disease activity; PhGA, physician global assessment of disease activity; SD, standard deviation; SJC, swollen joint count; SpA, spondyloarthritis; TJC, tender joint count; TNF, tumor necrosis factor; VAS, visual analogue score. Data presented as n (%) unless otherwise specified.

66 Pre-SpA INCEPTION COHORT Table 3. Clinical characteristics, laboratory and imaging results of the HLA-B27 positive and HLA-B27 negative FDRs HLA-B27 positive HLA-B27 negative (n=26) (n=25) p-value Demographics Male gender 15 (58) 10 (40) 0.211 Caucasian 25 (96) 25 (100) 0.327 Age, years, mean (SD) 25.3 (5.5) 26.3 (4.9) 0.396 2 BMI, kg/m , mean (SD) 24.1 (3.1) 22.9 (3.2) 0.248 Current smoker 8 (31) 6 (24) 0.753 4 Use of DMARD, TNF inhibitor 0 (0) 0 (0) - or corticosteroids ever

History Axial disease Back pain 12 (46) 17 (68) 0.119 Inflammatory back pain 6 (23) 5 (20) 0.791 Buttock pain 2 (10) 0 (0) 0.161 Good response to NSAIDs 4 (15) 1 (4) 0.176 Peripheral disease History of arthralgia 8 (46) 12 (32) 0.305 History of peripheral arthritis 1 (4) 0 (0) 0.327 History of enthesitis 0 (0) 0 (0) 0.327 History of dactylitis 0 (0) 0 (0) - Extra-articular disease History of psoriasis 2 (8) 0 (0) 0.161 History of IBD 0 (0) 1 (4) 0.308 History of urethritis/diarrhoea 0 (0) 0 (0) - History of uveitis 2 (8) 0 (0) 0.161

Family history Ankylosing spondylitis 26 (100) 25 (100) - Psoriatic arthritis 2 (8) 0 (0) 0.161 Psoriasis 1 (4) 1 (4) 0.978 Inflammatory bowel disease 1 (4) 5 (20) 0.076 Uveitis 1 (4) 0 (0) 0.327

Disease activity measurements PhGA, 0-100mm VAS, mean (SD) 5.88 (8.95) 7.52 (14.16) 0.898 PGA, 0-100mmVAS, mean (SD) 9.65 (13.29) 11.48 (19.00) 0.923 Patient nocturnal pain, 0-100mm VAS, mean (SD) 4.00 (7.47) 6.64 (18.35) 0.773 BASDAI, 0-10cm, mean (SD) 1.00 (1.05) 1.35 (1,51) 0.322 BASDAI >4 1 (4) 2 (8) 0.533 ASDAS-CRP, mean (SD) 1 (4) 0 (0) 0.947 ASDAS-CRP ≥1.3 6 (23) 7 (28) 0.690 BASFI, 0-10cm, mean (SD) 0.38 (0.60) 0.49 (0.93) 0.702

CHAPTER 4 67 Clinical examination modified Schober, cm, mean (SD) 5.85 (4.63) 6.51 (6.71) 0.212 modified Schober <4.5cm 3 (12) 9 (32) 0.079 Chest expansion, cm, mean (SD) 6.24 (2.74) 6.28 (2.23) 0.447 Chest expansion <3.6cm 1 (4) 1 (4) 0.978 BASMI, 0-10, mean (SD) 2.04 (0.77) 2.28 (0.74) 0.294 TJC > 0, 0-68 joints 5 (19) 3 (12) 0.482 SJC > 0, 0-66 joints 0 (0) 0 (0) - MASES, 0-13, mean (SD) 6 (23) 4 (16) 0.529 Dactylitis 0 (0) 0 (0) - Nail psoriasis 0 (0) 0 (0) - Psoriasis 0 (0) 0 (0) -

Laboratory CRP, mg/L, mean (SD) 2.37 (3.65) 2.37 (3.32) 0.763 CRP above 5 mg/L 3 (12) 3 (12) 0.960 ESR, mm/h, mean (SD) 6.68 (7.83) 6.76 (6.58) 0.502 ESR above 20 mm/h 2 (8) 2 (8) 1.000 Calprotectin, ng/mL, mean (SD) 309.39 (121.71) 381.50 (130.00) 0.048 Calprotectin above 715 ng/ml 0 (0) 0 (0) -

Imaging Low grade sacroiliitis on X-ray 2 (8) 1 (4) 0.556 SpA abnormalities of the lumbar spine on X-ray 0 (0) 0 (0) - Syndesmophytes of the cervical spine on X-ray 1 (4) 0 (0) 0.997 Inflammatory lesions of the SIJ on MRI 6 (24) 5 (20) 0.735 Inflammatory lesions of the spine on MRI 0 (0) 0 (0) -

SpA criteria mNY 0 (0) 0 (0) - ESSG 6 (23) 4 (16) 0.529 ASAS axSpA 5 (19) 4 (16) 0.765 ASAS peripheral SpA 0 (0) 0 (0) - CASPAR 0 (0) 0 (0) - ASAS, Assessment of SpondyloArthritis international Society; ASDAS, Ankylosing Spondylitis Disease Activity Score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; BMI, body mass index; CASPAR, ClASsification criteria for Psoriatic ARthritis; cm, centimeter; CRP, c-reactive protein; DMARD, disease modifying antirheumatic drugs; ESSG, European Spondyloarthropathy Study Group; ESR, erythrocyte sedimentation rate; FDR, first degree relatives; IBD, inflammatory bowel disease; HLA-B27, human leukocyte antigen-B27; kg, kilogram; MASES, Maastricht Ankylosing Spondylitis Enthesitis Score; m, meter; mm, millimeter; mNY, modified New York; MRI, magnetic resonance imaging; NSAIDs, non-steroidal anti-inflammatory drugs; PGA patient global assessment of disease activity; PhGA, physician global assessment of disease activity; SD, standard deviation; SIJ, sacroiliiac joints; SJC, swollen joint count; SpA, spondyloarthritis; TJC, tender joint count; TNF, tumor necrosis factor; VAS, visual analogue score. Data presented as n (%) unless otherwise specified.

68 Pre-SpA INCEPTION COHORT DISCUSSION

We report here the baseline demographic, clinical, laboratory, and imaging features of 51 individuals included in a prospective inception cohort study of FDRs of HLA-B27 positive AS patients. This study aims to investigate the earliest, pre-clinical phases of disease based on the concept that FDRs of HLA-B27 positive AS patients have a significantly increased risk to develop SpA. 4 A first major conclusion is that at baseline 33% of FDRs have clinical and/or imaging features that allow a classification of SpA according to the ASAS axSpA and/or ESSG classification criteria. These findings are perfectly in line with previous cross-sectional studies showing an increased SpA risk among FDRs. Remarkably, the recurrence rate in our study (33%) is even higher than in the reported studies (4.9-12% recurrence), while the mean age in our study was lower.13,18,34 Several factors may contribute to this higher recurrence rate. First, FDRs having clinical symptoms could be more willing to participate in our study than FDRs without symptoms, which may lead to channeling bias. Indeed, a substantial percentage of ‘seemingly healthy’ FDRs reported back pain when actively questioned for this symptom at baseline. The fact that these FDRs were not investigated for or diagnosed with axial SpA before inclusion in the study may be related either to the fact that the back pain symptoms were relatively mild or to ignorance of general physicians for these alarm symptoms. Second, we have defined SpA in this study using ASAS axSpA, ESSG, Amor, CASPAR, and mNY criteria whereas previous studies mostly used the mNY criteria. It is now well recognized that AS, defined by the mNY criteria, represents only a fraction of the total SpA spectrum. Supporting this explanation, none of the FDRs fulfilled the mNY criteria. Third, inclusion of MRI may have led to increased classification of FDRs. However, in our study only 2 out of 17 FDRs would not have been classified without MRI.35,36

In our study we classified FDRs according to the ASAS axSpA and/or ESSG classification criteria and did not diagnose SpA based on the rheumatologist’s opinion for several reasons. First, there are no diagnostic criteria for SpA. Second, a diagnosis based on expert’s opinion is subjective and has large intra- and inter-observer variation and therefore is not feasible in this cohort. A potential drawback of using SpA classification criteria, however, is that the a priori probability of fulfilling the ASAS axSpA, ESSG and/or Amor criteria is higher in these FDRs than in populations used to develop these criteria,29,35,37 because the study design implies that all participants have a positive family history and that half of them are HLA-B27 positive. Nevertheless, the fact that FDRs fulfilling the ASAS axSpA and/or ESSG classification criteria had significantly higher scores for disease activity than those not fulfilling the ASAS axSpA and/or ESSG classification criteria adds to the rightness of using these criteria for this particular study design.

Importantly, some key features of SpA including the presence of peripheral disease or

CHAPTER 4 69 extra-articular manifestations and increased inflammatory parameters were only rarely observed in FDRs and were also not different in those fulfilling the ASAS axSpA and/or ESSG classification criteria versus those who did not. Again, we cannot entirely exclude a selection bias here as we recruited ‘seemingly healthy’ FDRs and as axial symptoms may be more frequently unrecognized by FDRs and physicians than peripheral or extra- articular disease. Alternatively, this observation may suggest that axial disease precedes other SpA manifestations during disease development. Strikingly, 6 out of 38 (16%) FDRs not fulfilling the ASAS axSpA and/or ESSG classification criteria had imaging abnormalities suggestive for SpA, with one showing cervical syndesmophytes and 5 depicting sacroiliac BME on MRI. Moreover, we did observe abnormalities on the MRI SIJ but did not observe spine abnormalities on MRI. These individuals did not have any other clinical or biological signs of SpA. These imaging abnormalities might represent the subclinical phase. Future follow-up will learn if and which of these FDRs will evolve into the clinically established phase of SpA. We will be able to address the relevance of these observations in the near future by a) extending the cohort size which will allow to replicate the current findings, and b) prospective follow-up of these FDRs, especially focusing on those with imaging abnormalities only. A key question is whether these imaging abnormalities predict clinical signs and symptoms of SpA. A possibility is that a positive MRI, even when ‘highly suggestive of axial SpA’ is a feature that can be observed more frequently in ‘normal individuals’ than previously thought. Alternatively, part of the MRI abnormalities may indeed be the first signs of yet subclinical axial SpA.

Another intriguing finding was that 9 of 17 (53%) FDRs fulfilling SpA classification criteria were HLA-B27 negative, of whom 1 fulfilled both the ASAS axSpA and ESSG criteria, an additional 4 the ASAS axSpA and an additional 4 fulfilled the ESSG criteria. This proportion is higher than in earlier reports on FDRs (14-43%) in AS and non-radiographic axSpA,35,36,38 which indicated a lower recurrence risk in HLA-B27 negative FDRs. This could be due to the relative size of the current study. However, if confirmed in a larger sample set, further follow-up of these FDRs will allow us to determine if FDRs showing signs and symptoms of SpA will evolve to more active and severe disease, independently of HLA-B27 status or, alternatively, if the presence of HLA-B27 may promote exacerbation and persistence of subclinical pathology. Moreover, we could determine if genetic factors such as SNPs in ERAP- 1 and IL-23R may have a bigger contribution in the development of SpA than previously thought.

In conclusion, the data from this cohort suggest that a substantial proportion of FDRs of HLA-B27 positive AS patients between the age of 18-40 years do have clinical signs and symptoms and/or imaging abnormalities suggestive for SpA. Accordingly, 33% could already be classified as SpA. Extension of the sample size and follow-up of all subjects will allow us to disentangle the sequence of events that may lead to clinically manifest SpA and will allow us to determine which initial features may predict such a development.

70 Pre-SpA INCEPTION COHORT REFERENCES

1. Dougados M, Baeten D. Spondyloarthritis. Lancet of susceptibility loci in familial ankylosing spondylitis? 2011;377:2127-37. Arthritis Care Res (Hoboken) 2012;64:780-4. 2. Feldtkeller E, Khan MA, van der Heijde D, van der 15. Moller P, Vinje O, Dale K, Berg K, Kass E. Family studies Linden S, Braun J. Age at disease onset and diagnosis in Bechterew’s syndrome (ankylosing spondylitis). delay in HLA-B27 negative vs. positive patients with II. Prevalences of symptoms and signs in relatives ankylosing spondylitis. Rheumatol Int 2003;23:61-6. of HLAB27 negative probands. Scand J Rheumatol 3. Salvadorini G, Bandinelli F, Delle Sedie A, Riente L, 1984;13:11-4. Candelieri A, Generini S, et al. Ankylosing spondylitis: 16. van der Linden SM, Valkenburg HA, de Jongh BM, Cats 4 how diagnostic and therapeutic delay have changed A. The risk of developing ankylosing spondylitis in over the last six decades. Clin Exp Rheumatol HLA-B27 positive individuals. A comparison of relatives 2012;30:561-5. of spondylitis patients with the general population. 4. Sorensen J, Hetland ML. Decreases in diagnostic delay Arthritis Rheum 1984;27:241-9. are supported by sensitivity analyses. Ann Rheum Dis 17. Vinje O, Dale K, Moller P. Radiographic evaluation 2014;73:e45. of patients with Bechterew’s syndrome (ankylosing 5. Brandt HC, Spiller I, Song IH, Vahldiek JL, Rudwaleit M, spondylitis) and their first-degree relatives. Findings Sieper J. Performance of referral recommendations in in the spine and sacro-iliac joints and relations to non- patients with chronic back pain and suspected axial radiographic findings. Scand J Rheumatol 1985;14:119- spondyloarthritis. Ann Rheum Dis 2007;66:1479-84. 32. 6. Poddubnyy D, Vahldiek J, Spiller I, Buss B, Listing 18. Dernis E, Said-Nahal R, D’Agostino MA, Aegerter J, Rudwaleit M, et al. Evaluation of 2 screening P, Dougados M, Breban M. Recurrence of strategies for early identification of patients with spondylarthropathy among first-degree relatives of axial spondyloarthritis in primary care. J Rheumatol patients: a systematic cross-sectional study.Ann Rheum 2011;38:2452-60. Dis 2009;68:502-7. 7. Sieper J, Srinivasan S, Zamani O, Mielants H, Choquette 19. Chou CT, Lin KC, Wei JC, Tsai WC, Ho HH, Hwang CM, D, Pavelka K, et al. Comparison of two referral strategies et al. Study of undifferentiated spondyloarthropathy for diagnosis of axial spondyloarthritis: the Recognising among first-degree relatives of ankylosing spondylitis and Diagnosing Ankylosing Spondylitis Reliably (RADAR) probands. Rheumatology (Oxford) 2005;44:662-5. study. Ann Rheum Dis 2013;72:1621-7. 20. Reveille JD. The genetic basis of spondyloarthritis. Ann 8. van der Heijde D, Landewe R, Baraliakos X, Houben Rheum Dis 2011;70 Suppl 1:i44-50. H, van Tubergen A, Williamson P, et al. Radiographic 21. Ramiro S, Stolwijk C, van Tubergen A, van der Heijde findings following two years of infliximab therapy in D, Dougados M, van den Bosch F, et al. Evolution of patients with ankylosing spondylitis. Arthritis Rheum radiographic damage in ankylosing spondylitis: a 12 2008;58:3063-70. year prospective follow-up of the OASIS study. Ann 9. van der Heijde D, Landewe R, Einstein S, Ory P, Vosse Rheum Dis 2015;74:52-9. D, Ni L, et al. Radiographic progression of ankylosing 22. Sieper J, Rudwaleit M, Baraliakos X, Brandt J, spondylitis after up to two years of treatment with Braun J, Burgos-Vargas R, et al. The Assessment etanercept. Arthritis Rheum 2008;58:1324-31. of SpondyloArthritis international Society (ASAS) 10. van der Heijde D, Salonen D, Weissman BN, Landewe handbook: a guide to assess spondyloarthritis. Ann R, Maksymowych WP, Kupper H, et al. Assessment of Rheum Dis 2009;68 Suppl 2:ii1-44. radiographic progression in the spines of patients with 23. Ramiro S, van Tubergen A, Stolwijk C, van der Heijde ankylosing spondylitis treated with adalimumab for up D, Royston P, Landewe R. Reference intervals of spinal to 2 years. Arthritis Res Ther 2009;11:R127. mobility measures in normal individuals: the mobility 11. Hreggvidsdottir HS, Noordenbos T, Baeten DL. study. Ann Rheum Dis 2014. Inflammatory pathways in spondyloarthritis. Mol 24. Turina MC, Sieper J, Yeremenko N, Conrad K, Haibel Immunol 2014;57:28-37. H, Rudwaleit M, et al. Calprotectin serum level is 12. Yeremenko N, Paramarta JE, Baeten D. The an independent marker for radiographic spinal interleukin-23/interleukin-17 immune axis as progression in axial spondyloarthritis. Ann Rheum Dis a promising new target in the treatment of 2014;73:1746-8. spondyloarthritis. Curr Opin Rheumatol 2014;26:361- 25. Turina MC, Yeremenko N, Paramarta JE, De Rycke L, 70. Baeten D. Calprotectin (S100A8/9) as serum biomarker 13. Brown MA, Laval SH, Brophy S, Calin A. Recurrence risk for clinical response in proof-of-concept trials in axial modelling of the genetic susceptibility to ankylosing and peripheral spondyloarthritis. Arthritis Res Ther spondylitis. Ann Rheum Dis 2000;59:883-6. 2014;16:413. 14. Joshi R, Reveille JD, Brown MA, Weisman MH, Ward 26. van der Linden S, Valkenburg HA, Cats A. Evaluation of MM, Gensler LS, et al. Is there a higher genetic load diagnostic criteria for ankylosing spondylitis. A proposal

CHAPTER 4 71 for modification of the New York criteria. Arthritis Assessment of Spondyloarthritis International Society Rheum 1984;27:361-8. criteria for the classification of spondyloarthritis in early 27. Creemers MC, Franssen MJ, van’t Hof MA, Gribnau FW, spondyloarthritis clinics participating in the ESPERANZA van de Putte LB, van Riel PL. Assessment of outcome programme. Rheumatology (Oxford) 2014;53:353-60. in ankylosing spondylitis: an extended radiographic scoring system. Ann Rheum Dis 2005;64:127-9. 28. Rudwaleit M, Jurik AG, Hermann KG, Landewe R, van der Heijde D, Baraliakos X, et al. Defining active sacroiliitis on magnetic resonance imaging (MRI) for classification of axial spondyloarthritis: a consensual approach by the ASAS/OMERACT MRI group. Ann Rheum Dis 2009;68:1520-7. 29. Rudwaleit M, van der Heijde D, Landewe R, Listing J, Akkoc N, Brandt J, et al. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection. Ann Rheum Dis 2009;68:777-83. 30. Rudwaleit M, van der Heijde D, Landewe R, Akkoc N, Brandt J, Chou CT, et al. The Assessment of SpondyloArthritis International Society classification criteria for peripheral spondyloarthritis and for spondyloarthritis in general. Ann Rheum Dis 2011;70:25-31. 31. Dougados M, van der Linden S, Juhlin R, Huitfeldt B, Amor B, Calin A, et al. The European Spondylarthropathy Study Group preliminary criteria for the classification of spondylarthropathy. Arthritis Rheum 1991;34:1218-27. 32. Amor B, Dougados M, Mijiyawa M. [Criteria of the classification of spondylarthropathies]. Rev Rhum Mal Osteoartic 1990;57:85-9. 33. Taylor W, Gladman D, Helliwell P, Marchesoni A, Mease P, Mielants H, et al. Classification criteria for psoriatic arthritis: development of new criteria from a large international study. Arthritis Rheum 2006;54:2665-73. 34. J. DB, Polman A, de B-M. Hereditary factors in rheumatoid arthritis and ankylosing spondylitis. Ann Rheum Dis 1961;20:215-20. 35. Molto A, Paternotte S, Comet D, Thibout E, Rudwaleit M, Claudepierre P, et al. Performances of the Assessment of SpondyloArthritis International Society axial spondyloarthritis criteria for diagnostic and classification purposes in patients visiting a rheumatologist because of chronic back pain: results from a multicenter, cross-sectional study. Arthritis Care Res (Hoboken) 2013;65:1472-81. 36. Weber U, Hodler J, Jurik AG, Pfirrmann CW, Rufibach K, Kissling RO, et al. Assessment of active spinal inflammatory changes in patients with axial spondyloarthritis: validation of whole body MRI against conventional MRI. Ann Rheum Dis 2010;69:648-53. 37. van den Berg R, de Hooge M, van Gaalen F, Reijnierse M, Huizinga T, van der Heijde D. Percentage of patients with spondyloarthritis in patients referred because of chronic back pain and performance of classification criteria: experience from the Spondyloarthritis Caught Early (SPACE) cohort. Rheumatology (Oxford) 2013;52:1492-9. 38. Tomero E, Mulero J, de Miguel E, Fernandez-Espartero C, Gobbo M, Descalzo MA, et al. Performance of the

72 Pre-SpA INCEPTION COHORT 4

CHAPTER 4 73

Calprotectin (S100A8/9) as serum biomarker for clinical response in proof-of-concept trials in axial and peripheral spondyloarthritis

Maureen C. Turina1, Nataliya Yeremenko1,2, Jacqueline E. Paramarta1, Leen De Rycke1,3, Dominique L. Baeten1,2

1Department of Clinical Immunology and Rheumatology, Academic Medical Center/ University of Amsterdam, Amsterdam, The Netherlands; 2Laboratory of Experimental Immunology, Academic Medical Center/University of Amsterdam, Amsterdam, the Netherlands;

3 5 Department of Rheumatology and Clinical Immunology, University Medical Center Utrecht, Utrecht, The Netherlands

Arthritis Res Ther. 2014;16(5):413. ABSTRACT

Introduction Biomarkers complementing clinical evaluations may help to reduce the length and size of proof-of-concept (PoC) trials aimed to obtain quick go or no-go decisions in the clinical development of new treatments. We aimed to identify and validate serum biomarkers with a high sensitivity to change upon effective treatment in spondyloarthritis (SpA) PoC trials.

Methods The candidate biomarkers high sensitive-C-reactive protein (hs-CRP), interleukin-6 (IL-6), pentraxin-3 (PTX-3), alpha-2-macroglobulin (alpha-2-MG), matrix metalloproteinase-3 (MMP-3), calprotectin, and vascular endothelial growth factor (VEGF) were determined by enzyme-linked immunosorbent assay (ELISA) in healthy controls (n=20) and SpA patients before and after 2 weeks of infliximab (n=18) or placebo (n=19) treatment in cohort1. Clinical outcome was evaluated at week 12. Results were validated in ankylosing spondylitis (AS) with infliximab (cohort 2, n=21) and peripheral SpA with etanercept (cohort 3, n=20).

Results Serum levels of calprotectin, hs-CRP, PTX-3, VEGF (all p<0.001) and MMP-3 (p=0.062), but not IL-6 and alpha-2-MG, were increased in SpA versus healthy controls. Treatment with infliximab, but not placebo, significantly decreased calprotectin (p<0.001) and hs-CRP (p<0.001) levels, with a similar trend for MMP-3 (p=0.063). The Standardized Response Mean (SRM), which reflects the ability to detect changes over time, was high for calprotectin (-1.26), good for hs-CRP (-0.96) and moderate for MMP-3 (-0.52). Calprotectin and hs-CRP, but not MMP-3, were good biomarkers for treatment response in axial and peripheral SpA as evaluated and confirmed in cohort 2 and 3, respectively.

Conclusions Calprotectin and hs-CRP are good serum biomarkers with high sensitivity to change upon effective treatment at the group level in small-scale, short-term PoC trials in SpA.

76 CALPROTECTIN IN PoC TRIALS INTRODUCTION

The use of tumor necrosis factor (TNF) blockers has dramatically improved the clinical outcome and quality of life of patients with spondyloarthritis (SpA). Originally implemented in ankylosing spondylitis (AS) and psoriatic arthritis (PsA), there is now increasing evidence that they are equally effective in other subtypes such as non-radiographic axial SpA and non-AS, non-PsA peripheral SpA.1-6 Nevertheless, TNF blockade in SpA still has important limitations. First, contra-indications and side effects, especially in patients with severe comorbidity, preclude the use of TNF blockers in a significant proportion of SpA patients. Second, TNF blockade fails to fully control signs and symptoms of disease in a significant 5 proportion of patients, with only 15 to 20% reaching partial remission.1-6 Third, the currently available evidence suggests that TNF blockade fails to halt new bone formation and ankylosis.7 Finally, TNF blockade fails to induce long-lasting drug-free remission as disease relapses within a few months of treatment interruption in a large majority of patients.8,9

These limitations indicate that there is still an important unmet medical need for alternative treatment options in SpA. However, the success of TNF blockade also raises important medical and ethical hurdles to perform large, long-term, placebo-controlled trials with novel drugs. Additional strategies to obtain quick go and no-go signals in small short-term proof-of- concept (PoC) trials before proceeding to large long-term clinical studies are thus required. Such tools may comprise biomarkers which reliably show high sensitivity to change upon clinical response at the group level. A biomarker is a ‘characteristic that can be objectively measured and evaluated as an indicator of a normal biologic process, a pathophysiologic process, or a pharmacologic response to a therapeutic intervention’.10 If sensitive and reproducible, changes in biomarker levels during treatment may thus help to substantiate a genuine effect on specific biological processes in PoC trials.

Several biomarkers have previously been directly or indirectly associated with treatment response in SpA. C-reactive protein (CRP) is known to be elevated in active disease and to decrease upon effective treatment in SpA, but is less useful than in rheumatoid arthritis (RA) as two thirds of the SpA patients have normal CRP levels.11-14 Interleukin-6 (IL-6) is the main driver of CRP; serum IL-6 levels are significantly increased in active SpA and decrease upon clinical response after TNF blockade.15,16 Whereas CRP belongs to the short pentraxins, the pentraxin family also contains long pentraxins, including pentraxin-3 (PTX-3), which have been proposed as biomarkers of inflammation.17 Alpha-2-macroglobulin (alpha-2-MG) is another CRP-independent acute phase response protein that is used as serum biomarker.18

Besides acute phase response proteins, several potential biomarkers have been identified based on their involvement in the immunopathology of SpA. Matrix metalloproteinase-3 (MMP-3) levels are associated with peripheral arthritis, disease activity, progression of structural damage and response to TNF inhibition in SpA.19-21 Calprotectin, a heterodimer

CHAPTER 5 77 consisting of S100A8 and S100A9 (previously called myeloid-related protein [MRP] 8 and 14, respectively), is expressed and secreted during monocyte infiltration into inflamed tissues, including macrophage infiltration in SpA synovitis.22-24 Calprotectin serum levels are elevated in SpA, correlate well with disease activity and decrease upon TNF blockade.23 Finally, SpA synovitis is characterized by pronounced hypervascularity,25,26 which is at least in part mediated by vascular endothelial growth factor (VEGF). VEGF serum levels are elevated and correlate well with inflammation and disease activity in SpA.27,28

In this study, we aimed to systematically assess the value of these serum proteins as biomarkers of effective treatment in PoC trials with TNF blockers in SpA. More specifically, we aimed to identify which of these serum proteins can best discriminate between anti-TNF treatment and placebo as early as 2 weeks after start of treatment.

METHODS

Patients and samples Serum samples and clinical data for 20 healthy controls (HCs) and 78 SpA patients fulfilling the European Spondylarthropathy Study Group (ESSG)29 criteria were retrieved from our arthritis biobank for this analysis according to the study protocol as approved by the Medical Ethics Committee of the Academic Medical Center/University of Amsterdam (2013_057). The SpA population consisted of 3 separate cohorts, where cohort 2 and 3 are considered as independent validation cohorts. Cohort 1 consisted of 18 SpA patients with AS (n=8), PsA (n=8) and undifferentiated spondyloarthritis (USpA) (n=2), who received infliximab (5 mg/kg intravenously) and 19 SpA patients with AS (n=9), PsA (n=9) and USpA (n=1), who received placebo at week 0, 2 and 6.1 Cohort 2 consisted of 21 patients with AS according to the modified New York criteria, treated with infliximab (5mg/kg intravenously) at week 0, 2 and 6.30 Cohort 3 consisted of 20 peripheral SpA patients treated with etanercept (25 mg subcutaneously bi-weekly).31 None of the patients had been previously treated with a TNF blocker. Serum was obtained at baseline and week 2 for the first two cohorts, and at baseline and week 4 for cohort 3. The complete description of these cohorts and the response to treatment was reported previously.1, 30, 31

Serum biomarkers We determined the serum levels of the following candidate biomarkers by enzyme-linked immunosorbent assay (ELISA) as indicated by the manufacturers: high-sensitivity (hs)-CRP (Research & Diagnostic Systems, Inc. Minneapolis), IL-6 (Research & Diagnostics Systems, Inc. Minneapolis), PTX-3 (Research & Diagnostic Systems, Inc.), alpha-2-MG (Abcam, Cambridge, UK), MMP-3 (Biotrak, Amersham Pharmacia Biotech, Buckinghamshire, UK), VEGF (Research

78 CALPROTECTIN IN PoC TRIALS & Diagnostics Systems, Inc.), and calprotectin (Hycult Biotech, the Netherlands). All assays were performed in duplicate.

Statistical analysis Data were represented as mean (standard error of the mean) and compared using the parametric t-test (paired where appropriate) using the GraphPad Prism software program version 5.01. A p-value below 0.05 was considered as statistically significant. The standardized response mean (SRM) was calculated as the mean change in a parameter in a defined period of time divided by the standard deviation (SD) of that change. An SRM below 0.5, between 0.5 and 0.8, and above 0.8 indicates a poor, moderate and good potential to detect changes 5 over time, respectively.32 Power calculations to attain 90% power with a significance level of 0.05 were performed using the nQuery Advisor version 7.0 software program.

RESULTS

Baseline serum biomarkers in SpA versus HCs We first assessed which serum biomarkers were increased in active SpA compared to HCs using the baseline samples of cohort 1 (Figure 1), assuming that serum biomarkers that are increased in active disease could potentially be modulated and normalized by effective treatment. We thus used this strategy to select biomarkers of interest; it was not the aim to identify markers with a high diagnostic discriminative capacity between SpA and HCs. The serum levels of hs-CRP (16.44±2.36 µg/ml versus 0.84±0.27 µg/ml, p<0.001), PTX-3 (271.4 ± 35.0 pg/ml versus 46.0±23.6 pg/ml, p<0.001), calprotectin (1.76±0.13 µg/ml versus 0.65±0.10 µg/ml, p<0.001) and VEGF (237.0±30.3 pg/ml versus 16.7±5.7 pg/ml, p<0.001) were significantly higher in active SpA versus HCs, with a similar trend for MMP-3 (0.96±0.21 µg/ml versus 0.29±0.04 µg/ml, p=0.062). In contrast, levels of alpha-2-MG and IL-6 were not significantly different between active SpA and HCs. Elevated serum levels, defined as more than the mean + 2 SD of the serum levels of HCs, were observed in 86% of the SpA patients for calprotectin, 82% for VEGF, 81% for hs-CRP, 74% for MMP-3, and 59% for PTX-3.

Short-term modulation of serum biomarkers by infliximab in SpA We next assessed which of these biomarkers were significantly downregulated by infliximab but not placebo treatment as early as two weeks after initiation of therapy. We previously reported that both arms of this cohort were well randomized for disease activity (as judged by parameters such as patient’s and physician’s global assessment of disease activity and Bath ankylosing spondylitis disease activity index (BASDAI), and that there was a significant decrease in these parameters during treatment with infliximab but not placebo.1 Serum

CHAPTER 5 79 Figure 1. Serum biomarker levels in active spondyloarthritis (SpA) (n=37) versus healthy controls (HC) (n=20). Independent sample t-tests were performed. Data are represented as mean (standard error of the mean). *p<0.05. hs-CRP, high sensitive C-reactive protein; IL-6, interleukin-6; PTX-3, pentraxin-3; alpha-2-MG, alpha-2- macroglobulin; MMP-3, matrix metalloproteinase-3; VEGF, vascular endothelial growth factor.

levels of the investigated biomarkers were similar at baseline between the infliximab and placebo arms of cohort 1 (Figure 2). In the placebo group, none of the serum biomarkers changed significantly between baseline and week 2 (Figure 2). In contrast, we observed a significant decrease in serum levels of hs-CRP (p<0.001), IL-6 (p=0.047), and calprotectin

80 CALPROTECTIN IN PoC TRIALS (p<0.001) after two weeks of infliximab treatment, with a similar trend for MMP-3 (p=0.063) (Figure 2). Levels of PTX-3, alpha-2-macroglobulin and VEGF were not significantly modulated by infliximab over this period.

5

Figure 2. Changes in serum biomarkers between baseline (week 0) and week 2 of infliximab (IFX) (n=18) versus placebo (PLC) (n=19) treatment in active SpA. Paired t-tests were performed. Data are represented as mean (standard error of the mean), *p<0.05. hs-CRP, high sensitive C-reactive protein; IL-6, interleukin-6; PTX-3, pentraxin-3; alpha-2-MG, alpha-2-macroglobulin; MMP-3, matrix metalloproteinase-3; VEGF, vascular endothelial growth factor.

CHAPTER 5 81 Potential of serum biomarkers to change sensitively upon active treatment in SpA As the previous analyses identified calprotectin, hs-CRP, IL-6 and MMP-3 as the most promising biomarkers of treatment response, we next assessed their potential to change upon active treatment by using the SRM, which reflects the ability to detect changes over time.32 The SRM in the infliximab group was -1.26 for calprotectin, -0.96 for hs-CRP, -0.56 for IL-6, and -0.52 for MMP-3. In comparison, the SRM in the placebo group was -0.20 for calprotectin, -0.13 for hs-CRP, -0.05 for IL-6, and 0.23 for MMP-3. Thus, the delta SRM (SRM of active treatment minus SRM of placebo) was -1.06 for calprotectin, -0.83 for hs- CRP, -0.51 for IL-6, and -0.75 for MMP-3. As calprotectin and hs-CRP correlated moderately (r=0.648), we next aimed to confirm that an early change in calprotectin levels is superior to CRP as biomarker of clinically effective treatment by performing a power analysis on the data of cohort 1. Demonstration of a significant difference between infliximab and placebo treatment using changes in calprotectin levels over two weeks as outcome parameter with a power of 90% and a significance level of 0.05 would require 15 patients per group. A similar analysis with CRP would require 24 patients per group. Finally, we determined if a combination of changes in calprotectin and hs-CRP levels would be superior in treatment response compared to changes in calprotectin alone. Changes in calprotectin levels are independently associated to treatment response but when combining with changes to hs- CRP levels the association to treatment response was no longer significant.

Serum biomarkers of active treatment in AS Cohort 1 consisted of different SpA subtypes, mainly AS and PsA. To ascertain that the biomarker findings are not biased by the SpA subtype and that the data can be extrapolated to more homogenous SpA study populations, we performed two additional sets of analysis. First, we reanalyzed the biomarker data of cohort 1 separately for AS (n=8 in the infliximab group and n=9 in the placebo group) and PsA (n=8 in the infliximab group and n=9 in the placebo group). In the AS group, only hs-CRP (p=0.006) and calprotectin (p=0.028) decreased significantly after two weeks of infliximab treatment (Figure 3A). Decreases in IL-6 and MMP-3 levels did not reach significance and none of the biomarkers changed in the placebo arm (data not shown). The SRM was -0.98 for calprotectin and -1.39 for hs-CRP. For PsA patients, only calprotectin (p=0.003) decreased significantly after two weeks of infliximab treatment, with a similar but non-significant trend for hs-CRP (p=0.071) (Figure 3B). None of the biomarkers changed in the placebo arm (data not shown). The SRMs in PsA patients were -1.58 for calprotectin and -0.75 for hs-CRP.

Second, we performed a similar biomarker analysis in an independent cohort of patients with AS without peripheral arthritis or enthesitis (n=20) receiving infliximab at week 0, 2 and 6 (cohort 2). As shown in Figure 3C, hs-CRP (p=0.001) and calprotectin (p=0.019) serum levels decreased significantly after two weeks of infliximab treatment. MMP-3 (p=0.093) showed a similar but non-significant trend in this cohort. The SRMs was -0.56 for calprotectin and

82 CALPROTECTIN IN PoC TRIALS -0.59 for hs-CRP. Taken together, these data indicate that changes in calprotectin and hs-CRP serum levels reflect well the response to active treatment with infliximab in AS.

5

Figure 3. Changes in serum biomarker levels between baseline (week 0) and week 2 of infliximab treatment in the ankylosing spondylitis (AS) patients (n=8) (A) and psoriatic arthritis (PsA) patients (n=8) (B) of cohort 1 as well as in the AS patients from cohort 2, who did not have peripheral disease (n=21) (C). Paired t-tests were performed. Data are represented as mean (standard error of the mean). *p<0.05. hs-CRP, high sensitive CRP; MMP- 3, matrix metalloproteinase-3.

Serum biomarkers of treatment response in peripheral SpA The previous analyses suggest that there may be some differences in serum biomarkers between AS and PsA. To confirm and extend these data and to ascertain that these biomarkers do not only reflect response to infliximab but also other TNF inhibitors, we analyzed changes of serum biomarkers between baseline and week 4 of etanercept treatment in patient with peripheral SpA (cohort 3). As shown in Figure 4, we again observed a significant decrease in serum levels of calprotectin (p=0.032), hs-CRP (p=0.035) and MMP-3 (p=0.045). In this cohort, the SRMs were -0.55 for calprotectin, -0.54 for hs-CRP, and -0.51 for MMP-3. Furthermore, changes in calprotectin levels were not influenced by the presence of psoriasis.

CHAPTER 5 83 Figure 4. Changes in serum biomarkers between baseline (week 0) and week 4 of etanercept treatment (n=20) in peripheral SpA. Paired t-tests were performed. Data are represented as mean (standard error of the mean). *p<0.05. hs-CRP, high sensitive C-reactive protein; MMP-3, matrix metalloproteinase-3.

Calprotectin as a biomarker in CRP-negative patients Both calprotectin and CRP are not highly specific to SpA and are elevated in several diseases. In SpA, CRP levels are not elevated in two thirds of the patients. Furthermore, the correlation between CRP and calprotectin is modest in SpA (r=0.634), suggesting that calprotectin may provide more or additional information in comparison with CRP. We analyzed whether calprotectin is of additional value by determining the sensitivity to change upon treatment effect in CRP-negative patients. We therefore stratified for CRP status (positive or negative), using the threshold for normal values of our local laboratory.

For cohort 1, the delta calprotectin levels were -0.30±0.21 μg/ml in CRP-negative patients versus -0.90±0.13 μg/ml in CRP-positive patients, with a corresponding SRM of -0.79 and -1.60 in CRP-negative and CRP-positive patients, respectively. For cohort 2, the delta calprotectin levels were -0.19±0.14 μg/ml in CRP-negative patients versus -0.35±0.15 μg/ml in CRP-positive patients, with a corresponding SRM of -0.38 and -0.81 in CRP-negative and CRP-positive patients, respectively. For cohort 3, the delta serum calprotectin levels was -0.10±0.11 μg/ml in CRP-negative patients versus -0.30±0.18 μg/ml in CRP-positive patients, with a corresponding SRM of -0.35 and -0.66 in CRP-negative and CRP-positive patients, respectively.

DISCUSSION

Tissue inflammation is one of the major features of both axial and peripheral SpAbut remains difficult to quantify and monitor in an objective and reliable way in clinical research. The most commonly used biomarkers for inflammation in SpA are magnetic resonance imaging (MRI) and CRP. MRI is able to visualize inflammatory lesions such as synovitis, enthesitis and bone marrow oedema in axial and peripheral SpA and these lesions decrease or even disappear upon effective treatment.33,34 Accordingly, MRI is now used as measure for inflammation in clinical trials in AS and axial SpA. However, even in patients with high

84 CALPROTECTIN IN PoC TRIALS clinical disease activity MRI scores are often low, which makes it difficult to detect significant pre- and post-treatment changes in small size PoC trials.35 The second widely used measure of inflammation in SpA is CRP. CRP levels do correlate at the group level with clinical disease activity, decrease upon effective treatment and are included in the ankylosing spondylitis disease activity score (ASDAS).36 As for MRI, however, an important issue is that baseline serum levels of CRP are elevated only in a fraction of patients with active SpA, hs-CRP may therefore perform better than classical CRP measurements.13,37

In the current study we explored the value of different candidate serum biomarkers of clinical response in SpA, thereby focusing on the specific situation of small, short-term PoC trials. We aimed to identify serum proteins that are significantly modulated by effective 5 treatment with TNF blockers but not by placebo as early as 2 weeks after treatment initiation. In line with the previously discussed data, we found that hs-CRP is a useful biomarker of inflammation in this context as, despite not being elevated in all patients at baseline, it did rapidly and significantly decrease in both axial and peripheral SpA treated with either infliximab or etanercept. The high SRM and the absence of changes in the placebo-treated patients confirm its value as biomarker of inflammation in PoC trials. Interestingly, serum levels of IL-6, the major driver of CRP, did not appear to be a reliable marker of treatment response in our study. Moreover, two recent clinical trials with IL-6-receptor blockers failed to demonstrate clinical efficacy in AS.38,39 Taken together, these data suggest that, even though CRP may be used as biomarker, the IL-6-CRP axis is not a pivotal inflammatory pathway in SpA.

The major finding of the study is that serum levels of calprotectin appeared tobethe most robust and reliable marker of treatment response in both axial and peripheral SpA, even outperforming hs-CRP in some analyses. For example, a power calculation in the infliximab cohort indicated that the use of calprotectin rather than hs-CRP as biomarker for inflammation would allow to a significant reduction in the size of the cohort. In contrast to CRP, there is also clear evidence that calprotectin directly reflects key inflammatory processes in SpA. Calprotectin is produced by myeloid cells and neutrophils and is secreted when these cells transmigrate through the endothelium to the inflamed tissues. Elevated levels of calprotectin as measured in synovial fluid in arthritis or in the faeces in inflammatory bowel disease thus reflect local tissue infiltration by myeloid cells.40,41 We and others have demonstrated that calprotectin is released locally in SpA synovitis, that the degree of synovial infiltration with myeloid cells and neutrophils reflect disease activity in peripheral SpA, and that effective treatment of peripheral SpA is associated with a rapid and robust decrease of calprotectin-positive cells in the inflamed synovium.22-24, 42 In contrast, much less is known about calprotectin expression in axial disease. Taken together, these data indicate that secretion of calprotectin may be a good reflection of local migration and activation of inflammatory innate immune cells in target tissues of SpA. Accordingly, a decrease in these calprotectin levels upon treatment is likely to reflect a genuine biological effect on chronic

CHAPTER 5 85 tissue inflammation.

Although we used only TNF blockers as effective drugs in the present study, we have now evidence that the biomarker value of calprotectin is not restricted to this sole mode of action. In a recent PoC study with the monoclonal anti-IL17A antibody, secukinumab, we demonstrated that both CRP and S100A8 and A9 levels decreased rapidly in the treated patients.35 Supporting the concept that calprotectin may have additive value compared to CRP, changes between baseline and week 6 levels of S100A8 and A9, but not CRP, correlated with assessment of spondyloarthritis 20 (ASAS20) and ASAS40 responses at week 6. Moreover, baseline levels of S100A8 and S100A9 but not CRP had a high sensitivity to change upon clinical response to treatment in this trial.

The combination of calprotectin and hs-CRP as biomarkers for clinical response, does not give any additional value compared to calprotectin alone. In contrast to calprotectin and hs-CRP, the other candidate biomarkers did not perform that well in our studies. In line with previous reports,16,20,28,43 serum levels of markers such as VEGF and MMP-3 did decrease upon effective anti-inflammatory therapy but the treatment effects were either smaller or less consistent across different SpA subtypes and cohorts in comparison with calprotectin and hs-CRP. This may also be due to the specific setting of the present study where we aimed to detect very early changes (at week 2) in biomarker levels.

It is important to emphasize here that our data certainly do not imply that serum biomarkers such as calprotectin or hs-CRP should replace clinical outcome parameters. Indeed, clinical outcomes (including composite indices such as BASDAI and ASDAS) can also show significant and rapid changes in anti-TNF treated patients. However, the correlation between a serum biomarker such as calprotectin and clinical outcomes such as BASDAI is significant but weak (r=0.325), indicating that both measures do not cover exactly the same aspects ofthe disease. Clinical outcome parameters may also be more sensitive to placebo responses and thus combination of clinical outcome measures with serum biomarkers may help to obtain a consistent picture which can support go or no-go decisions in proof-of-concept trials. A second aspect to emphasize is that calprotectin, similarly to CRP, is not exclusively elevated in SpA but also in many other inflammatory conditions.44 Co-morbidity may influence serum levels of calprotectin and CRP in SpA. Nevertheless, we demonstrated here that effective treatment still induces rapid, robust and consistent changes of these serum biomarkers which are not observed during placebo treatment.

Our study has two important limitations. First, we used a candidate approach based on a limited set of potential biomarkers emerging from literature and/or insights in tissue inflammation in SpA. However, a more systematic approach using proteomics orgene expression arrays allowing identification of novel biomarkers therefore certainly remains warranted. For example, an unbiased microarray analysis of peripheral blood cells could identify a set of genes which expression was closely correlated to disease activity in AS; one

86 CALPROTECTIN IN PoC TRIALS of these genes codes for LIGHT and serum levels of LIGHT were confirmed to correlate with CRP and erythrocyte sedimentation rate (ESR).45 Second, we focused on the use of serum biomarkers in a very specific setting, namely the measurement of a biological impact on inflammation in PoC trials. The set-up and size of the studies used in our analyses do not allow determination of whether changes in calprotectin and hs-CRP levels not only reflect treatment response at the group level but also allow to distinguish at the individual level responders from non-responders to TNF blockade. Moreover, it remains unknown if serum biomarkers such as calprotectin may also reflect other important clinical features of SpA. As, for example, both CRP and VEGF serum levels correlate with osteoproliferation in axial SpA,46,47 it would be interesting to explore the value of calprotectin in similar settings. Finally, 5 we did not explore yet the value of calprotectin as biomarker of early disease, including non- radiographic axial SpA and patients presenting with early inflammatory back pain.

CONCLUSIONS

In summary, good biomarkers are needed in the development of novel therapies in SpA to provide quick go and no-go signals for PoC trials. Our data indicate that calprotectin and hs- CRP perform well as serum biomarkers as these biomarkers have a high sensitivity to change upon clinical effective treatment and could be of use in the development of new treatments.

CHAPTER 5 87 REFERENCES

1. Van den Bosch F, Kruithof E, Baeten D, et al. Randomized 2009;61:1484-1490. double-blind comparison of chimeric monoclonal 13. Poddubnyy DA, Rudwaleit M, Listing J, et al. antibody to tumor necrosis factor alpha (infliximab) Comparison of a high sensitivity and standard C versus placebo in active spondylarthropathy. Arthritis reactive protein measurement in patients with Rheum 2002;46:755-765. ankylosing spondylitis and non-radiographic axial 2. Braun J, Brandt J, Listing J, et al. Treatment of active spondyloarthritis. Ann Rheum Dis 2010;69:1338- ankylosing spondylitis with infliximab: a randomised 1341. controlled multicentre trial. Lancet 2002;359:1187- 14. Visvanathan S, Wagner C, Marini JC, et al. 1193. Inflammatory biomarkers, disease activity and 3. Mease PJ, Goffe BS, Metz J, et al. Etanercept in the spinal disease measures in patients with ankylosing treatment of psoriatic arthritis and psoriasis: a spondylitis after treatment with infliximab. Ann randomised trial. Lancet 2000;356:385-390. Rheum Dis 2008;67:511-517. 4. Paramarta JE, De Rycke L, Heijda TF, et al. Efficacy 15. Gratacos J, Collado A, Filella X, et al. Serum cytokines and safety of adalimumab for the treatment of (IL-6, TNF-alpha, IL-1 beta and IFN-gamma) in peripheral arthritis in spondyloarthritis patients ankylosing spondylitis: a close correlation between without ankylosing spondylitis or psoriatic arthritis. serum IL-6 and disease activity and severity. Br J Ann Rheum Dis 2013;72:1793-1799. Rheumatol 1994;33:927-931. 5. Sieper J, van der Heijde D, Dougados M, et al. 16. Pedersen SJ, Hetland ML, Sorensen IJ, et al. Circulating Efficacy and safety of adalimumab in patients with levels of interleukin-6, vascular endothelial growth non-radiographic axial spondyloarthritis: results of a factor, YKL-40, matrix metalloproteinase-3, and total randomised placebo-controlled trial (ABILITY-1). Ann aggrecan in spondyloarthritis patients during 3 years Rheum Dis 2013;72:815-22. of treatment with TNFalpha inhibitors. Clin Rheumatol 6. Van der Heijde D, Kivitz A, Schiff MH, et al.; ATLAS 2010;29:1301-1309. Study Group. Efficacy and safety of adalimumab 17. Garlanda C, Bottazzi B, Bastone A, et al. Pentraxins in patients with ankylosing spondylitis: results of at the crossroads between innate immunity, a multicenter, randomized, double-blind, placebo- inflammation, matrix deposition, and female fertility. controlled trial. Arthritis Rheum 2006;54:2136-2146. Annu Rev Immunol 2005;23:337-366. 7. Van der Heijde D, Landewe R, Baraliakos X, et al.; 18. Surrall KE, Bird HA, Dixon JS. Et al. Prealbumin and Ankylosing Spondylitis Study for the Evaluation alpha 2-macroglobulin as potential indices of disease of Recombinant Infliximab Therapy Study Group. activity in different arthritides. Clin Rheumatol Radiographic findings following two years of infliximab 1987;6:64-69. therapy in patients with ankylosing spondylitis. 19. Yang C, Gu J, Rihl M, et al. Serum levels of matrix Arthritis Rheum 2008;58:3063-3070. metalloproteinase 3 and macrophage colony- 8. Baraliakos X, Listing J, Brandt J, et al. Clinical response stimulating factor 1 correlate with disease activity in to discontinuation of anti-TNF therapy in patients ankylosing spondylitis. Arthritis Rheum 2004;51:691- with ankylosing spondylitis after 3 years of continuous 699. treatment with infliximab. Arthritis Res Ther 20. Vandooren B, Kruithof E, Yu DT, et al. Involvement 2005;7:R439-R444. of matrix metalloproteinases and their inhibitors in 9. Paramarta JE, Heijda TF, Baeten DL. Fast relapse upon peripheral synovitis and down-regulation by tumor discontinuation of tumour necrosis factor blocking necrosis factor alpha blockade in spondylarthropathy. therapy in patients with peripheral spondyloarthritis. Arthritis Rheum 2004;50:2942-2953. Ann Rheum Dis 2013;72:1581-2. 21. Maksymowych WP, Landewe R, Conner-Spady B, et al. 10. De Gruttola VG, Clax P, De Mets DL, et al. Considerations Serum matrix metalloproteinase 3 is an independent in the evaluation of surrogate endpoints in clinical predictor of structural damage progression in trials. summary of a National Institutes of Health patients with ankylosing spondylitis. Arthritis Rheum workshop. Control Clin Trials 2001;22:485-502. 2007;56:1846-1853. 11. Benhamou M, Gossec L, Dougados M. Clinical 22. Kane D, Roth J, Frosch M, et al. Increased perivascular relevance of C-reactive protein in ankylosing synovial membrane expression of myeloid-related spondylitis and evaluation of the NSAIDs/coxibs’ proteins in psoriatic arthritis. Arthritis Rheum treatment effect on C-reactive protein.Rheumatology 2003;48:1676-1685. (Oxford) 2010;49:536-541. 23. De Rycke L, Baeten D, Foell D, et al. Differential 12. De Vries MK, van Eijk IC, van der Horst-Bruinsma IE, expression and response to anti-TNFalpha treatment et al. Erythrocyte sedimentation rate, C-reactive of infiltrating versus resident tissue macrophage protein level, and serum amyloid a protein for patient subsets in autoimmune arthritis.J Pathol 2005;206:17- selection and monitoring of anti-tumor necrosis factor 27. treatment in ankylosing spondylitis. Arthritis Rheum

88 CALPROTECTIN IN PoC TRIALS 24. Kruithof E, De Rycke L, Vandooren B, et al.; OMERACT patients with ankylosing spondylitis. Ann Rheum Dis Special Interest Group on Synovial Analysis in Clinical 2009;68:18-24. Trials. Identification of synovial biomarkers of 37. Navarro-Compán V, van der Heijde D, Combe B, et response to experimental treatment in early-phase al. Value of high-sensitivity C-reactive protein for clinical trials in spondylarthritis. Arthritis Rheum classification of early axial spondyloarthritis: results 2006;54:1795-1804. from the DESIR cohort. Ann Rheum Dis 2013;72:785- 25. Baeten D, Demetter P, Cuvelier C, et al. Comparative 786. study of the synovial histology in rheumatoid arthritis, 38. Sieper J, Porter-Brown B, Thompson L, et al. spondyloarthropathy, and osteoarthritis: influence Assessment of short-term symptomatic efficacy of disease duration and activity. Ann Rheum Dis of tocilizumab in ankylosing spondylitis: results of 2000;59:945-953. randomised, placebo-controlled trials. Ann Rheum 26. Baeten D, Kruithof E, De Rycke L, et al. Diagnostic Dis Published Online First: 13 June 2013. doi:10.1136/ classification of spondylarthropathy and rheumatoid annrheumdis-2013-203559. arthritis by synovial histopathology: a prospective 39. Sieper J, Inman RD, Badalamenti S, et al. Sarilumab study in 154 consecutive patients. Arthritis Rheum for the treatment of ankylosing spondylitis: results 5 2004;50:2931-2941. of a phase 2, randomized, double-blind, placebo- 27. Drouart M, Saas P, Billot M, et al. High serum vascular controlled, international study (ALIGN) [abstract].Ann endothelial growth factor correlates with disease Rheum Dis 2012;71(Suppl 3):111. activity of spondylarthropathies. Clin Exp Immunol 40. Vogl T, Ludwig S, Goebeler M, et al. MRP8 and 2003;132:158-162. MRP14 control microtubule reorganization during 28. Pedersen SJ, Sorensen IJ, Garnero P, et al. ASDAS, transendothelial migration of phagocytes. Blood BASDAI and different treatment responses and their 2004;104:4260-8. relation to biomarkers of inflammation, cartilage and 41. Foell D, Wittkowski H, Ren Z, et al. Phagocyte-specific bone turnover in patients with axial spondyloarthritis S100 proteins are released from affected mucosa and treated with TNFalpha inhibitors. Ann Rheum Dis promote immune responses during inflammatory 2011;70:1375-1381. bowel disease. J Pathol 2008;216:183-192. 29. Dougados M, van der Linden S, Juhlin R, et al. 42. Baeten D, Kruithof E, De Rycke L, et al. Infiltration of The European Spondylarthropathy Study Group. the synovial membrane with macrophage subsets The European Spondylarthropathy Study Group and polymorphonuclear cells reflects global disease preliminary criteria for the classification of activity in spondyloarthropathy. Arthritis Res Ther spondylarthropathy. Arthritis Rheum 1991;34:1218- 2005;7:R359-R369. 1227. 43. Appel H, Janssen L, Listing J, et al. Serum levels of 30. Franco Salinas G, De Rycke L, Barendregt B. et al. biomarkers of bone and cartilage destruction and Anti-TNF treatment blocks the induction of T cell- new bone formation in different cohorts of patients dependent humoral responses. Ann Rheum Dis with axial spondyloarthritis with and without tumor 2013;72:1037-1043. necrosis factor-alpha blocker treatment. Arthritis Res 31. Kruithof E, De Rycke L, Roth J, Mielants, et al. Ther 2008;10:R125. Immunomodulatory effects of etanercept on 44. 44. Schiopu A, Cotoi OS: S100A8 and S100A9: peripheral joint synovitis in the spondylarthropathies. DAMPs at the crossroads between innate immunity, Arthritis Rheum 2005;52:3898-3909. traditional risk factors, and cardiovascular disease. 32. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for Mediators Inflamm 2013, 2013: 828354. interpreting changes in health status. Med Care 45. 45. Haroon N, Tsui FW, O’Shea FD, et al. From gene 1989;27(3 Suppl):S178-89. expression to serum proteins: biomarker discovery in 33. Sieper J, Rudwaleit M, Baraliakos X, et al. The ankylosing spondylitis. Ann Rheum Dis 2010;69:297- Assessment of SpondyloArthritis international Society 300. (ASAS) handbook: a guide to assess spondyloarthritis. 46. 46. Poddubnyy D, Haibel H, Listing J, et al. Baseline Ann Rheum Dis 2009;68 Suppl 2:ii1-44. radiographic damage, elevated acute-phase 34. Lambert RG, Salonen D, Rahman P, et al. Adalimumab reactant levels, and cigarette smoking status significantly reduces both spinal and sacroiliac joint predict spinal radiographic progression in early axial inflammation in patients with ankylosing spondylitis: spondylarthritis. Arthritis Rheum 2012;64:1388-1398. a multicenter, randomized, double-blind, placebo- 47. 47. Poddubnyy D, Conrad K, Haibel H, et al. Elevated controlled study. Arthritis Rheum 2007;56:4005-4014. serum level of the vascular endothelial growth 35. Baeten D, Baraliakos X, Braun J, et al. Anti-interleukin- factor predicts radiographic spinal progression in 17A monoclonal antibody secukinumab in treatment patients with axial spondyloarthritis. Ann Rheum Dis of ankylosing spondylitis: a randomised, double-blind, Published Online First: 16 August 2013. doi: 10.1136/ placebo-controlled trial. Lancet 2013;382:1705-13. annrheumdis-2013-203824. 36. Lukas C, Landewé R, Sieper J, et al. Assessment of SpondyloArthritis international Society. Development of an ASAS-endorsed disease activity score (ASDAS) in

CHAPTER 5 89

A psychometric analysis of outcome measures in peripheral spondyloarthritis

Maureen C. Turina1*, Sofia Ramiro1 *, Dominique L. Baeten1, Philip Mease2, Jacqueline E. Paramarta1, In-Ho Song3, Aileen L. Pangan3, Robert Landewé1

1Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and immunology Center, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; 2Swedish Medical Center & University of Washington, Seattle, Washington, United States; 6 3AbbVie Inc., North Chicago, Illinois, United States

*Equal contributions

Ann Rheum Dis. 2015 Aug 5;1–6 ABSTRACT

Objectives To assess the discriminatory capacity of various outcome measures and response criteria in patients with peripheral SpondyloArthritis (pSpA).

Methods Data originated from two randomized controlled trials, ABILITY-2 and Tnf Inhibition in Peripheral SpondyloArthritis (TIPES). Continuous outcome measures included patient’s global assessment (PGA)/physician’s global assessment of disease (PhGA), C-reactive protein (CRP), tender joint counts (TJC)/swollen joint counts (SJC), Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and the Ankylosing Spondylitis Disease Activity Score (ASDAS). Dichotomous response criteria included Peripheral SpondyloArthritis Response Criteria (PSpARC), American College of Rheumatology (ACR), ASDAS and BASDAI response criteria. The capacity to discriminate between adalimumab and placebo groups was assessed by standardised mean differences (SMD) for continuous variables, and Pearson’s χ2 for dichotomous response criteria.

Results Within each trial, the composite indices for axial SpA assessment, ASDAS-CRP (SMD: -0.63 and -0.89 in ABILITY-2 and the TIPES trial, respectively) and BASDAI (SMD: –0.50 and -0.73), and the single-item measures PGA (SMD: -0.47 and -1.12) and PhGA (SMD: -0.64 and -0.87) performed better than other single-item measures, such as CRP (SMD: -0.18 and -0.53), SJC or TJC. In general, the PSpARC and ACR response criteria discriminated better than ASDAS and BASDAI response criteria.

Conclusions The axial SpA-specific ASDAS-CRP and BASDAI, but also PGA and PhGA, demonstrated good discriminatory ability in patients with pSpA. The pSpA-specific pSpARC response criteria and the RA-specific ACR response criteria also discriminated well. To fully capture typical pSpA manifestations, it may be worth developing new pSpA-specific indices with better performance and face validity.

92 ANALYSIS OF OUTCOME MEASURES IN pSpA INTRODUCTION

Spondyloarthritis (SpA) is a spectrum of diseases with several subtypes with overlapping clinical, radiographic and genetic characteristics.1 Recently, the Assessment in SpondyloArthritis international Society (ASAS) has developed classification criteria for SpA, based on the predominant clinical manifestation, as either axial SpA (axSpA), presenting with chronic back pain, or peripheral SpA (pSpA), presenting with arthritis, enthesitis or dactylitis.1,2 Several tumour necrosis factor inhibitors (TNFi) have been approved for axSpA (both non-radiographic axSpA (nr-axSpA) and Ankylosing Spondylitis (AS)), as well as for Psoriatic Arthritis (PsA), although not for pSpA.3-7 Two randomized controlled trials (RCTs) have been performed to assess the efficacy of TNFi in pSpA.8,9 Adalimumab (ADA) was effective in both trials, which had different primary endpoints because no composite 6 measures or response criteria had been previously validated in patients with pSpA. In the ABILITY-2 trial, a new composite outcome measure, the Peripheral SpondyloArthritis Response Criteria (PSpARC)40, was developed as the primary endpoint.9 In the Tnf Inhibition in PEripheral SpondyloArthritis (TIPES) trial, the improvement in patient’s global assessment of disease (PGA) was chosen as the primary endpoint.8

Therefore, there exists a need to identify discriminant outcome measures for pSpA. Several efficacy variables were used in the two RCTs, including measures developed specifically for AS, such as the Ankylosing Spondylitis Diseases Activity Score (ASDAS) and the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI); and measures developed specifically for rheumatoid arthritis (RA, also applied in PsA), including the American College of Rheumatology (ACR)20/50/70. However, the performance of all these outcome measures in pSpA is unknown, and we hoped to determine which measures best reflect disease activity and clinical response, because the success of future therapeutic clinical trials depends not only on a well-defined patient population, but also on the availability of valid outcome measures and response criteria. Taking into account the Outcome Measures in Rheumatology Clinical Trials (OMERACT) guidance,10,11 we compared the discriminative properties of outcome measures in pSpA. The aim of this study was to assess the sensitivity to change and discriminatory aspects of outcome measures and response criteria in pSpA.

METHODS

Patient population The analysis included data from two double-blind placebo (PBO)-controlled RCTs, ABILITY-2 and TIPES,8,9 which evaluated the efficacy and safety of ADA in active patients with pSpA. ABILITY-2 (NCT01064856) included patients fulfilling the ASAS pSpA criteria2 who did

CHAPTER 6 93 not have PsA or AS. The primary endpoint was PSpARC40 at week 12, defined as ≥40% improvement from baseline in PGA and patient global pain, and ≥40% improvement in any of the following: (1) swollen joint count (SJC) and tender joint count (TJC); (2) enthesitis count; or (3) dactylitis count.

The TIPES trial (EUDRACT 2008-006885-27) included patients with pSpA, fulfilling the European Spondyloarthropathy Study Group (ESSG) criteria and/or the Amor criteria,12,13 without PsA or AS. The primary endpoint was the change in PGA at week 12.

Outcome measures and response criteria The performance of the following outcome measures in assessing disease activity and treatment response after 12 weeks was evaluated: PGA, patient global pain, physician’s global assessment of disease activity (PhGA), ASDAS-CRP, BASDAI, SJC, TJC and C-reactive protein (CRP).

The patient global pain in the past week, PGA, and PhGA of current disease activity were recorded on a 0-100 mm visual analogue scale (VAS). The ASDAS-CRP, originally developed for AS, includes questions pertaining to axial and peripheral symptoms, PGA and CRP.14 Disease activity was classified as follows: <1.3, inactive disease; 1.3 to <2.1, moderate disease; 2.1 to ≤3.5, high disease; and >3.5, very high disease.15 The BASDAI, also developed for AS, consists of questions mainly for axial and peripheral complaints, measured on a 0-10 cm VAS.

Clinical response criteria assessed were PSpARC40/50/70, ASDAS-major improvement (ASDAS-MI, change in ASDAS≥2.0), ASDAS-clinically important improvement (ASDAS- CII, change in ASDAS≥1.1), ASDAS-inactive disease (ASDAS-ID, ASDAS<1.3),15 BASDAI50 (improvement of ≥50% in BASDAI score), BASDAI ≥2 units (improvement of ≥2 units),16 and ACR20/50/70.

Since the TIPES trial did not capture patient global pain, enthesitis or dactylitis counts, a modified PSpARC40 was determined, which included 40% improvements from baseline in PGA, patient global pain and SJC66 and TJC68. Patient global pain was calculated as the mean of BASDAI components #3 (joint pain/swelling) and #4 (enthesitis), because these showed the highest correlation to patient global pain in the ABILITY-2 trial (Spearman’s coefficient=0.6). Modified ACR20/50/70 criteria were derived for the TIPES study, where patient global pain was calculated as above.

Statistical analysis Analyses were performed for all outcome measures, for the two RCTs in the ‘as observed’ population, without imputation for rare missing data. The RCTs were separately analysed because this allowed results from one trial to be used to confirm results from the other, and

94 ANALYSIS OF OUTCOME MEASURES IN pSpA also because there were differences in the outcome measures collected and the inclusion criteria.

First, we evaluated whether levels assessed by these measures could discriminate between two states of disease activity. Since no gold standard is available to define disease activity states in pSpA, patients (in each trial) were assigned to (arbitrarily determined) states of low disease activity and high disease activity based on the PhGA and PGA at baseline of treatment (<40 vs ≥60 mm, excluding patients with values in between, in order to increase separation). In subgroups with low disease activity and high disease activity, the standardised mean difference (SMD) was calculated as the difference between the group means divided by the pooled standard deviation (SD). The SMD has no units, which facilitates comparison across disease measures. An SMD with higher absolute value indicates better discriminatory ability. 6

For the continuous outcome measures, the sensitivity to change of each outcome measure for detecting improvement from baseline to week 12 was determined by comparing the adjusted standardised means of change from baseline to week 12 for all measures for both treatment groups separately. Adjusted standardised mean changes from baseline to week 12 were obtained for each continuous outcome measure (dependent), stratified for treatment (ADA and PBO) and adjusted for baseline values of the corresponding outcome measure (covariate), using analysis of co-variance (ANCOVA). The following formula was used for standardising: ‘disease activity outcome change (week 12-baseline) divided by SD of that outcome at baseline’. These standardised mean changes reflect sensitivity to change of an outcome measure within a treatment group (ADA vs PBO).

We then evaluated whether responses assessed by these measures could discriminate between ADA treatment and PBO treatment by determining SMDs, which reflect the capacity of continuous outcomes measures to discriminate between change under ADA and change under PBO. Furthermore, the t-score and the Guyatt’s effect size (ES) were determined for discriminatory capacity and sensitivity to change, respectively. Guyatt’s ES is the mean change in the treatment group divided by the SD of the PBO group, and relates the magnitude of the effect (the ‘signal’) to the magnitude of the non-specific change (the ‘noise’). Guyatt’s ES of 0.2, 0.5 and 0.8 represent small, medium and large effect size, respectively.17 A higher t-score indicates a better discriminatory ability within the same trial. The discriminatory ability of the dichotomous response criteria was determined by Pearson’s c2 (chi-square) or Fisher’s exact test (if n<5). With a constant number of observations per outcome measure, a higher c2 indicates better discriminatory ability. Analyses were performed using SPSS V.20 (SPSS, Chicago, Illinois, USA).

CHAPTER 6 95 RESULTS

Demographics and disease characteristics In total, 205 patients were included: 165 from ABILITY-2 and 40 patients from the TIPES trial. The primary results have been reported previously.8,9 Patients with pSpA were more often female and the mean age was around 40 years. Half of them were HLA-B27 positive and the mean number of swollen joints was 5-6 (see online supplementary table S1). Of note, a high proportion had a history of enthesitis, and in ABILITY-2, dactylitis was not common. About half of them were using disease-modifying antirheumatic drugs (DMARDs), most often methotrexate or sulfasalazine, in equal proportions. Overall, the outcome measures were similar at baseline between the ADA and the PBO groups in both studies.

Discrimination between disease activity states In the ABILITY-2 subgroups with low disease activity versus high disease activity at baseline based on PhGA, the SMD was highest for ASDAS-CRP (1.16) followed by BASDAI (1.13), patient global pain (1.03), and BASDAI #3 (1.00) (Table 1). In the TIPES trial, the SMD was highest for BASDAI #1 (2.66), PGA (2.01), ASDAS-CRP (1.84) and BASDAI (1.75). In the ABILITY-2 subgroups with low disease activity and high disease activity based on PGA, the SMD was highest for patient global pain (4.46) followed by ASDAS-CRP (2.08), BASDAI #4 (1.62) and BASDAI (1.61) (data not shown). In the TIPES trial, the SMD was highest for BASDAI (2.22) followed by ASDAS-CRP (2.18), patient global pain (1.89), and PhGA (1.71).

Sensitivity to change of continuous outcome measures All measures showed an improvement after 12 weeks of treatment in both trials, with greater improvements observed in patients on ADA versus PBO treatment (Figure 1). Across both studies, the outcome measures with greatest sensitivity in detecting change from baseline after 12 weeks of treatment were PhGA, PGA, patient global pain, ASDAS-CRP, BASDAI and SJC. The ranking of the adjusted standardised means of change from baseline of the measures differed slightly between the trials. In ABILITY-2, the largest changes from baseline were observed in PhGA, patient global pain, PGA, and ASDAS-CRP. In the TIPES study, the largest changes from baseline were observed in ASDAS-CRP, PGA, PhGA and patient global pain.

Discriminatory aspects of continuous outcome measures In ABILITY-2, PhGA had the highest SMD (-0.64), (as well as Guyatt’s ES and t-scores, data not shown), followed by ASDAS-CRP (-0.63), patient global pain (-0.50), BASDAI (-0.50), TJC78 (-0.50), and PGA (-0.47) (Table 2). In the TIPES trial, PGA had the highest SMD (-1.12), (and

96 ANALYSIS OF OUTCOME MEASURES IN pSpA Table 1. Validation constructs: Discrimination between low disease activity and high disease activity according to physician’s global assessment (PhGA) at baseline ABILITY-2 TIPES Low (<40) High (≥60) SMD Low (<40) High (≥60) SMD mean (SD) mean (SD) mean (SD) mean (SD) n=15 n=84 n=6 n=12 PGA, 0-100 mm VAS 55.3 (17.0) 71.0 (15.7) 0.99 45.8 (13.4) 75.2 (15.1) 2.01 Patient global pain, 53.7 (14.8) 69.7 (15.5) 1.03 34.4 (21.0) 58.2 (18.5) -0.04 0-100 mm VAS ASDAS-CRP, 0-10 cm VAS 2.3 (0.8) 3.2 (0.8) 1.16 1.9 (0.7) 3.4 (0.9) 1.84 BASDAI total (0-10 cm 4.2 (1.9) 6.0 (1.5) 1.13 3.4 (1.5) 6.2 (1.7) 1.75 VAS) BASDAI #1 4.4 (1.9) 6.6 (2.1) 0.89 2.8 (1.1) 7.2 (1.9) 2.66 BASDAI #2 2.7 (2.7) 4.3 (3.0) 0.53 3.4 (2.6) 5.9 (3.5) 0.77 6 BASDAI #3 5.4 (2.2) 7.1 (1.6) 1.00 3.9 (2.6) 6.2 (2.6) 0.87 BASDAI #4 4.3 (2.8) 6.5 (2.1) 0.97 3.0 (2.3) 5.5 (2.8) 0.94 BASDAI #5 4.9( 2.9) 6.1 (2.4) 0.50 4.9 (2.5) 7.1 (2.2) 1.00 BASDAI #6 3.4 (2.8) 5.6 (2.8) 0.78 2.6 (1.9) 5.2 (3.5) 0.83 BASDAI mean#5,#6 #5,#6 4.1 (2.7) 5.8 (2.3) 0.71 3.7 (2.1) 6.2 (2.7) 1.00 SJC76a 5.1 (5.1) 7.5 (8.1) 0.31 2.5 (2.7) 4.2 (3.4) 0.52 TJC78b 8.3 (4.9) 14.8 (15.7) 0.45 5.7 (3.8) 9.0 (6.5) 0.57 CRP (mg/L) 7.3 (8.9) 13.0 (20.7) 0.29 1.6 (1.1) 16.7 (31.0) 0.59 aIn TIPES, SJC was 0-66 joints. bTJC 0-68 joints and patient global pain were derived from the mean of BASDAI #3 and #4. SD, standard deviation; SMD, standardised mean difference; #, number; VAS, visual analog scale; PGA, patient global assessment; PhGA, physician global assessment; ASDAS-CRP, Ankylosing Spondylitis Disease Activity Score based on C-reactive protein; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; SJC, swollen joint count; TJC, tender joint count; CRP, C-reactive protein;

Guyatt’s ES and t-scores, data not shown), followed by patient global pain (-0.93); ASDAS- CRP (-0.89), PhGA (-0.87), some single-item components of BASDAI and BASDAI itself (-0.73). SJC discriminated well in the TIPES, but not the ABILITY-2 trial. Since the discriminatory performance of outcome measures had not been previously investigated in the pSpA population, the effects of level of low disease activity and high disease activity at baseline on performance were measured. The measures performed similarly, independently of level of disease activity at baseline (data not shown). However, in ABILITY-2, but not TIPES, the discriminatory ability of ASDAS-CRP, BASDAI, PGA, PhGA and TJC78 was enhanced in the subgroups with higher disease activity at baseline defined by BASDAI, compared to the subgroup with lower disease activity.

CHAPTER 6 97 ABILITY-2 ADA PBO

PhGA

Patient global pain

PGA

ASDAS-CRP

BASDAI #3

BASDAI #4

BASDAI

SJC 76

BASDAI #5

TJC 78 BASDAI #1

BASDAI #6

BASDAI #2 CRP

-2.0 -1.5 -1.0 -0.5 0.0 0.5 TIPES

ASDAS-CRP

PGA PhGA

Patient global pain

SJC 66

BASDAI #5

BASDAI

TJC 68

BASDAI #3 CRP

BASDAI #6

BASDAI #4

BASDAI #1

BASDAI #2

-2.0 -1.5 -1.0 -0.5 0.0 0.5 adjusted standardized mean of change from baseline

Figure 1. Sensitivity to change of continuous outcome measures. Adjusted standardised mean changes from baseline for all outcome measures, adjusted for baseline values of variables. Outcomes are ranked independently for each trial in descending order based on the adjusted standardised means of change for the active treatment group. For TIPES, patient global pain was derived from the mean of BASDAI #3 and #4.

98 ANALYSIS OF OUTCOME MEASURES IN pSpA Table 2. Discrimination of disease activity measurements between patients on ADA versus PBO ABILITY-2 TIPES Mean Δ from Mean Δ from SMD Mean Δ from Mean Δ from SMD BL ADA BL PBO BL ADA BL PBO (SD), N=82 (SD), N=81 (SD), N=19 (SD), N=19 PGA, 0-100 mm VAS -28.0 (26.0)b -16.2 (24.5) d -0.47 -31.0 -5.9 (21.4) -1.12 (23.3) Patient global pain, -29.3 (24.6) -17.0 (24.4) d -0.50 -21.9 (26.9) -0.6 (17.5) -0.93 0-100 mm VAS PhGA, 0-100 mm VAS -32.7 (22.5) -18.2 (22.9) -0.64 -19.8 (19.5) -4.1 (16.4) -0.87 ASDAS-CRP, 0-10 cm VAS -1.1 (1.1)a -0.5 (0.9)c -0.63 -1.5 (1.2) -0.6 (0.8) -0.89 BASDAI, 0-10 cm VAS -2.1 (2.3) -1.0 (2.2)d -0.50 -1.9 (2.6) -0.3 (1.5) -0.73 BASDAI #1 -1.9 (2.6) -1.0 (2.6) -0.36 -1.2 (3.6) -0.5 (2.3) -0.20 BASDAI #2 -1.2 (3.0) -0.0 (2.4) -0.42 -1.7 (3.3) -0.8 (2.3) -0.33 BASDAI #3 -2.9 (2.9) -1.7 (2.9) -0.40 -2.6 (3.5) -0.0 (1.5) -0.96 6 BASDAI #4 -2.7 (2.9) -1.2 (2.7) -0.54 -1.8 (2.7) -0.1 (2.8) -0.62 BASDAI #5 -2.3 (2.8) -1.1 (2.7) -0.44 -2.8 (3.3) -0.7 (2.1) -0.77 BASDAI #6 -1.8 (2.8) -1.2 (2.7) -0.21 -1.2 (3.3) 0.0 (2.2) -0.43 BASDAI mean #5 and #6 -2.1 (2.6) -1.2 (2.5) -0.35 -2.0 (3.1) -0.3 (1.9) -0.64 SJC 76e -3.6 (4.3) -3.1 (5.6) -0.10 -2.5(4.1) -0.4 (1.8) -0.67 TJC 78f -6.0 (8.7) -1.8 (8.4) -0.50 -1.8 (9.2) 1.7 (6.5) -0.45 CRP (mg/L) -5.8 (18.7)b -2.9 (11.0)d -0.18 -5.7 (12.4) 4.0 (22.9) -0.53 a N=80, b N=81, c N=77, d N=80. e In TIPES, SJC was 0-66 joints, f In TIPES, TJC was 0-68 joints and patient global pain was derived from the mean of BASDAI #3 and #4. Δ, change; #, number; SD, standard deviation; SMD, standardised mean difference; BL, baseline; ADA, adalimumab; PBO, placebo; PGA, patient global assessment; PhGA, physician global assessment; ASDAS-CRP, Ankylosing Spondylitis Disease Activity Score based on C-reactive protein; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; SJC, swollen joint count; TJC, tender joint count; CRP, C-reactive protein; VAS, visual analog scale. Observed data.

Discriminatory ability of categorical clinical response criteria Among the response criteria used in both studies, ACR20/50 and PSpARC40/50 performed comparatively well in differentiating between ADA and PBO treatment with c2 (in ABILITY-2 and the TIPES trial respectively) for ACR20 of 16.05 and 11.79; for ACR50, 13.66 and 8.58; for PSpARC40, 8.18 and 8.58; for PSpARC50, 13.46 and 7.13 (Table 3). PSpARC70 showed significant discriminatory ability in ABILITY-2, although not in the TIPES trial (c2 13.49 and 3.26, respectively). Among the AS-specific measures used in ABILITY-2 and the TIPES study respectively, ASDAS-ID (c2 7.17 and 10.13), ASDAS-CII (c2 9.44 and 6.91) and BASDAI50 (c2 10.90 and 7.13) showed significant discriminatory activity across trials. As expected, the TIPES trial reached less discrimination compared to ABILITY-2.

CHAPTER 6 99 Table 3. Discrimination between patients on adalimumab versus placebo at week 12 using clinical response criteria ABILITY-2 TIPES ADA PBO ADA PBO n (%) n (%) Pearson’s n (%) n (%) Pearson’s N=82 N=80 χ2 P-value N=19 N=19 χ2 P-value PSpARC40 33 (41)a 16 (20) 8.18 0.004 7 (37) 0 (0) 8.58 0.008 PSpARC50 29 (36)a 9 (11) 13.46 <0.001 6 (32) 0 (0) 7.13 0.020 PSpARC70 19 (24)b 3 (4) 13.49 <0.001 3 (16) 0 (0) 3.26 0.230 ASDAS-MI 18 (23)b 5 (6)c 8.04 0.005 5 (26) 2 (11) 1.58 0.405 ASDAS-CII 35 (44)b 16 (21)c 9.44 0.002 12 (63) 4 (21) 6.91 0.020 ASDAS-ID 27 (34)b 12 (15)d 7.17 0.007 8 (42) 0 (0) 10.13 0.003 BASDAI50 35 (43) 15 (19) 10.90 0.001 8 (42) 1 (5) 7.13 0.019 BASDAI≥2 39 (48) 24 (30) 5.25 0.022 7 (37) 2 (11) 3.64 0.124 ACR 20 47 (57) 21 (26) 16.05 <0.001 9 (47) 0 (0) 11.79 0.001 ACR 50 28 (34) 8 (10) 13.66 <0.001 7 (37) 0 (0) 8.58 0.008 ACR 70 15 (18) 2 (3) 10.75 0.001 4 (21) 0 (0) 4.47 0.105 a N=81, b N=80, c N=77, dN=78. χ2, Chi-square; ADA, adalimumab; PBO, placebo; PSpARC, peripheral SpondyloArthritis response criteria; ASDAS-MI, Ankylosing Spondylitis Disease Activity Score-major improvement; ASDAS-CII, ASDAS- clinically important improvement; ASDAS-ID, ASDAS-inactive disease; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; ACR, American College of Rheumatology. Observed data. For TIPES, modified PSpARC and ACR were calculated.

DISCUSSION

In conducting trials, as well as in monitoring patients in clinical practice, there is a need to define the optimal measures for disease activity and clinical response. This is acknowledged as an important research focus for pSpA, given the growing awareness and diagnosis of this disease and the need for new therapies. In our assessment of the performance and hierarchy of outcomes in two independent RCTs, we have found that among the status measures evaluated for disease activity, ASDAS-CRP, BASDAI, PGA, patient global pain and PhGA consistently had both, the highest sensitivity to change from baseline and the highest level of discriminatory ability. Concerning the response criteria (to be used in RCTs), in both trials ACR20 and PSpARC50/70 performed best in terms of discrimination. Previously, a similar analysis in AS18 determined that ASDAS-CRP performed extremely well compared to other outcome measures with respect to sensitivity to change and discrimination.

The two RCTs used different inclusion criteria: ABILITY-2 used the ASAS criteria, whereas TIPES used the ESSG/Amor criteria, resulting in slight differences in the patient populations. Also, the trials partly assessed different outcome measures: PSpARC, ACR and patient global pain were assessed in ABILITY-2 but not TIPES. Therefore, we analyzed them independently. Importantly, this allowed the validation of findings from one population in the other, thus adding robustness. As there is no gold standard for low disease activity and high disease activity in pSpA, we artificially constructed states of low disease activity or

100 ANALYSIS OF OUTCOME MEASURES IN pSpA high disease activity based on two external constructs: PhGA and PGA. Regardless of the external construct used, the AS-specific indices ASDAS-CRP and BASDAI showed the best performance in both trials. The relatively better performance of the ASDAS and BASDAI indices, which were originally developed for AS, may be because both measures include aspects of peripheral joints (the presence of peripheral swelling; BASDAI also includes a question regarding enthesitis).

Interestingly, PGA and PhGA, which are non-disease specific measures, performed as well as ASDAS-CRP and BASDAI (axSpA-specific measures). This finding was consistent across all analyses, supported by the fact that these four outcome measures showed the best discrimination between treatment groups. Notably, these non-disease specific measures were used in the first trials with TNFi in SpA,19, 20 when disease-specific outcome measures had not yet been developed. Possibly a similar cycle of outcome measure development may 6 be required for pSpA. Our data strongly suggest that the perceptions of patients as well as physicians about disease activity are thus far not captured by existing disease activity indices and that the use of axSpA-specific disease measures inherently lacks face validity. The sensitivity and discriminatory ability of almost all these measures was increased in subgroups of patients with higher disease activity at baseline in the ABILITY-2 trial.

Among the response criteria evaluated, the PSpARC and the ACR criteria performed relatively well in both trials. The performance of the ACR20/50, which are used in RA and PsA, in the pSpA population, may be attributed to the overlapping arthritic symptoms between RA and pSpA. Although the ACR response criteria appeared to perform better than the PSpARC, unlike the PSpARC, these do not capture all manifestations of pSpA symptoms, which may limit their usefulness for patients with pSpA. In addition, the TIPES trial did not collect data on patient global pain, enthesitis and dactylitis and therefore the PSpARC was calculated using the PGA, patient global pain (calculated as mean of BASDAI questions 3 and 4), SJC and TJC. The performance of the axSpA-specific response criteria, ASDAS-MI and ASDAS–CII, and BASDAI50, was worse than ACR20 and PSpARC50/70. This is most likely because the cut-off levels for ASDAS-MI/CII and BASDAI50 were obtained and tested in populations with axSpA, and not in populations with pSpA. Our analyses suggest that certain disease activity states are specific to the population in which they have been validated. In other words, it is questionable if ASDAS/BASDAI response criteria should be used in populations with pSpA, despite their very acceptable psychometric characteristics in this regard.

To our knowledge, this is the first study specifically looking into outcome measures in pSpA. A limitation of this study is that due to the different outcomes assessed in the trials, patient global pain, pSpARC40/50/70, and ACR20/50/70 were retroactively derived for TIPES. The differences in performance of some measures between the two studies may have been influenced by smaller sample size of the TIPES trial. The strengths of this study include the reasonably high number of patients, the controlled prospective design, the availability of a

CHAPTER 6 101 placebo group for comparison, and the inclusion of many existing outcome measures and response criteria.

In conclusion, the continuous composite outcome measures ASDAS-CRP and BASDAI, as well as the single-item measures, PGA and PhGA, performed consistently well in both pSpA trials, and better than other single-item measures such as CRP and TJC, in detecting change from baseline, and in discriminating between active and placebo treatment. To fully capture typical pSpA manifestations such as enthesitis and dactylitis, it may be worthwhile to develop new composite measures, specific for pSpA, as the performance of PGA and PhGA in this analysis suggests that important parts of the patient’s and physician’s perceptions of disease activity are not yet captured by the current constructs. Regarding the response criteria, our results suggest the use of the disease-specific PSpARC and non-specific ACR criteria in future clinical trials because they represent multiple facets of pSpA disease (face validity), include patient’s and physician’s assessments (face validity), and performed well in both RCTs (discrimination) in comparison to other response criteria evaluated.

102 ANALYSIS OF OUTCOME MEASURES IN pSpA REFERENCES

1. Dougados M, Baeten D. SpondyloArthritis. 1991;34(10):1218-27. Lancet 2011;377(9783):2127-37. 13. Amor B, Dougados M, Mijiyawa M. [Criteria of 2. Rudwaleit M, van der Heijde D, Landewe R, the classification of spondylarthropathies].Rev et al. The Assessment of SpondyloArthritis Rhum Mal Osteoartic 1990;57(2):85-9. International Society classification criteria 14. Lukas C, Landewe R, Sieper J, et al. Development for peripheral SpondyloArthritis and for of an ASAS-endorsed disease activity score SpondyloArthritis in general. Ann Rheum Dis (ASDAS) in patients with ankylosing spondylitis. 2011;70(1):25-31. Ann Rheum Dis 2009;68(1):18-24. 3. Brandt J, Haibel H, Cornely D, et al. Successful 15. Machado P, Landewe R, Lie E, et al. Ankylosing treatment of active ankylosing spondylitis Spondylitis Disease Activity Score (ASDAS): with the anti-tumor necrosis factor alpha defining cut-off values for disease activity monoclonal antibody infliximab. Arthritis states and improvement scores. Ann Rheum Rheum 2000;43(6):1346-52. Dis 2011;70(1):47-53. 4. Mease PJ, Goffe BS, Metz J, et al. Etanercept 16. Braun J, Pham T, Sieper J, et al. International 6 in the treatment of psoriatic arthritis ASAS consensus statement for the use of and psoriasis: a randomised trial. Lancet anti-tumour necrosis factor agents in patients 2000;356(9227):385-90. with ankylosing spondylitis. Ann Rheum Dis 5. Sieper J, van der Heijde D, Dougados M, et al. 2003;62(9):817-24. Efficacy and safety of adalimumab in patients 17. Cohen J, editor. Statistical Power Analysis for with non-radiographic axial SpondyloArthritis: the Behavioral Sciences. . Hillsdale NJ: Erlbaum, results of a randomised placebo-controlled trial 1988. (ABILITY-1). Ann Rheum Dis 2013;72(6):815-22. 18. van der Heijde D, Braun J, Dougados M, et 6. van der Heijde D, Kivitz A, Schiff MH, et al. Sensitivity and discriminatory ability of al. Efficacy and safety of adalimumab in the Ankylosing Spondylitis Disease Activity patients with ankylosing spondylitis: results Score in patients treated with etanercept of a multicenter, randomized, double-blind, or sulphasalazine in the ASCEND trial. placebo-controlled trial. Arthritis Rheum Rheumatology (Oxford) 2012;51(10):1894-905. 2006;54(7):2136-46. 19. Van den Bosch F, Kruithof E, Baeten D, et al. 7. Landewe R, Braun J, Deodhar A, et al. Efficacy Effects of a loading dose regimen of three of certolizumab pegol on signs and symptoms infusions of chimeric monoclonal antibody to of axial SpondyloArthritis including ankylosing tumour necrosis factor alpha (infliximab) in spondylitis: 24-week results of a double-blind spondyloarthropathy: an open pilot study. Ann randomised placebo-controlled Phase 3 study. Rheum Dis 2000;59(6):428-33. Ann Rheum Dis 2014;73(1):39-47. 20. Van Den Bosch F, Kruithof E, Baeten D, et 8. Paramarta JE, De Rycke L, Heijda TF, et al. Randomized double-blind comparison al. Efficacy and safety of adalimumab for of chimeric monoclonal antibody to tumor the treatment of peripheral arthritis in necrosis factor alpha (infliximab) versus SpondyloArthritis patients without ankylosing placebo in active spondylarthropathy. Arthritis spondylitis or psoriatic arthritis. Ann Rheum Rheum 2002;46(3):755-65. Dis 2013;72(11):1793-9. 9. Mease P, Sieper J, van den Bosch F, et al. Randomized controlled trial of adalimumab in patients with non-psoriatic peripheral spondyloarthritis. Arthritis Rheum 2015;67(4):914-23. 10. Boers M, Brooks P, Strand CV, et al. The OMERACT filter for Outcome Measures in Rheumatology. J Rheumatol 1998;25(2):198-9. 11. Boers M, Kirwan JR, Wells G, et al. Developing core outcome measurement sets for clinical trials: OMERACT filter 2.0. J Clin Epidemiol 2014;67(7):745-53. 12. Dougados M, van der Linden S, Juhlin R, et al. The European Spondylarthropathy Study Group preliminary criteria for the classification of spondylarthropathy. Arthritis Rheum

CHAPTER 6 103 Supplemental Table S1. Baseline characteristics of the study populations in ABILITY-2 and TIPES – ITT ABILITY-2 TIPES Characteristics ADA (N=84) PBO(N=81) ADA (N=20) PBO (N=20) Demographics Age, years 42.5 (10.8) 38.5 (12.8) 41.5 (12.8) 44.4 (11.1) Disease duration, years 7.7 (7.9) 6.6 (6.3) 7.9 (9.3) 6.7 (6.2) Gender (male), n (%) 36 (42) 39 (48) 9 (45) 12 (60) Disease Characteristics HLA-B27 positive, n (%) 56 (67) 44 (56) 11 (55) 5 (25) IBD (past/present), n (%) 5 (6) 3 (4) 1 (5) 3 (15) PGA , 0-100 mm VAS 65.2 (15.2) 66.4 (15.9) 65.8 (18.1) 68.4 (17.4) Patient global pain, 0-100 mm VAS 64.3 (14.0) 65.6(15.9) 59.6 (21.5) 57.9 (16.4) PhGA , 100 mm VAS 60.3 (15.5) 57.0 (15.0) 47.8 (11.8) 57.0 (12.6) ASDAS-CRP 0-10 cm VAS 2.9 (0.8) 3.1 (0.8) 2.9 (1.0) 3.2 (0.9) BASDAI, 0-10 cm VAS 5.7 (1.7) 5.6 (1.6) 5.5 (2.3) 6.0 (1.4) BASDAI #1 6.0 (2.1) 5.8 (2.2) 5.7 (2.9) 6.5 (2.0) BASDAI #2 4.3 (3.1) 4.0 (2.8) 5.0 (3.3) 6.0 (2.9) BASDAI #3 6.6 (2.0) 6.7 (1.7) 6.3 (2.4) 5.9 (2.2) BASDAI #4 6.1 (2.4) 5.9 (2.2) 5.6 (2.9) 5.6 (2.2) BASDAI #5 5.7 (2.6) 5.8 (2.3) 5.9 (3.0) 6.7 (1.9) BASDAI #6 4.9 (2.8) 5.2 (2.8) 4.1 (3.1) 5.1 (3.3) BASDAI mean #5, #6 5.3 (2.4) 5.5 (2.3) 5.0 (2.9) 5.9 (2.3) SJC 6.1 (5.6)a 7.3 (8.0)a 4.3 (4.2)b 2.5 (1.9)b TJC 13.0 (12.8)c 13.6 (16.1)c 9.4 (8.2)d 10.6 (5.9)d Enthesitis (past/present), n (%) 48 (57) 57 (70) 13 (65) 14 (70) Dactylitis (past/present), n (%) 23 (27) 27 (33) 4 (20) 1 (5) CRP, mg/l 9.7 (17.7) 10.7 (15.7) 7.8 (13.3) 13.5 (26.4) Concomitant therapies Concomitant NSAIDs, n (%) 57 (68) 65 (80) 13 (65) 14 (70) Concomitant DMARD, n (%) 38 (45) 41 (51) 14 (70) 11 (55) Concomitant MTX, n (%) 23 (27) 25 (31) 5 (25) 6 (30) Concomitant SSZ, n (%) 20 (24) 27 (33) 7 (35) 4 (20) All values are the mean (SD), unless otherwise specified. †2 patients missing. ‡5mg/l is upper limit of normal in both trials. #, number; ITT, Intention to treat; ADA, adalimumab; PBO, placebo; HLA-B27, human leukocyte antigen B27; IBD, Inflammatory Bowel Disease; PGA, patient global assessment; PhGA, physician global assessment; VAS, visual analogue scale; ASDAS-CRP, Ankylosing Spondylitis Disease Activity Score based on C-reactive protein; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; SJC, swollen joint count; TJC, tender joint count; CRP, C-reactive protein; NSAID, nonsteroidal anti-inflammatory drug; DMARD, disease modifying antirheumatic drug; MTX, methotrexate; SSZ, sulfasalazine. aSJC 0-76 joints, bSJC 0-66 joints, cTJC 0-78 joints, dTJC 0-68 joints. For TIPES, patient global pain was derived from the mean of BASDAI #3 and #4.

104 ANALYSIS OF OUTCOME MEASURES IN pSpA 6

CHAPTER 6 105

Calprotectin serum level is an independent marker for radiographic spinal progression in axial spondyloarthritis

Maureen C. Turina1, Joachim Sieper2, Nataliya Yeremenko1,3, Kristina Conrad2, Hildrun Haibel2, Martin Rudwaleit4, Dominique Baeten1,3, Denis Poddubnyy2

1Department of Clinical Immunology and Rheumatology, Academic Medical Center/ University of Amsterdam, Amsterdam, the Netherlands; 2Rheumatology, Med. Department I, Campus Benjamin Franklin, Charité 7 Universitätsmedizin Berlin, Berlin, Germany; 3Laboratory of Experimental Immunology, Academic Medical Center/University of Amsterdam, Amsterdam, the Netherlands; 4Endokrinologikum Berlin, Berlin, Germany

Ann Rheum Dis. 2014;73(9):1746–8. In previous works, we demonstrated that markers of systemic inflammation (elevated C-reactive protein [CRP] and erythrocyte sedimentation rate [ESR]) are independently associated with radiographic spinal progression over two years in axial spondyloarthritis (axSpA) in the German spondyloarthritis inception cohort (GESPIC).1 We also found previously that calprotectin, which is secreted during monocyte infiltration into inflamed tissues, and thus, directly reflects a potentially important pathophysiological mechanism in SpA,2,3 and which was found to be associated with structural joint damage in psoriatic arthritis4 and in rheumatoid arthritis,5 is clearly elevated in SpA as compared to healthy controls, and decreases rapidly and consistently upon effective treatment,2 although in another work, no differences in serum calprotectin between ankylosing spondylitis patients and healthy controls could be found.6 We aimed here to assess whether serum calprotectin levels are predictive for progression of structural damage in the spine in axSpA.

Seventy-six patients (mean age 38.0±11.5 years, mean symptom duration 4.6±2.8 years, 66% males, 82% HLA-B27-positives) with definite axSpA (n=63 fulfilling the modified New York criteria for ankylosing spondylitis and n=13 patients with axSpA not fulfilling the radiographic part of the modified New York criteria, but with at least one syndesmophyte in the cervical or lumbar spine) from GESPIC,1,7 were selected for this analysis. Detailed description of the cohort and radiographs scoring has been reported elsewhere.1,7-9 Radiographic spinal progression was defined as (1) worsening of the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) by ≥2 units, and (2) development of a new syndesmophyte or progression of existing syndesmophytes (formation of a bridging syndesmophyte) after 2 years. Levels of serum calprotectin were determined by ELISA (Hycult Biotech, the Netherlands).

Baseline calprotectin serum level was significantly higher in patients with mSASSS worsening (n=15) versus those without mSASSS worsening (n=61): mean±SD 0.68±0.21, median 0.69 μg/mL versus mean±SD 0.48±0.26, median 0.42 μg/mL, respectively, p=0.005 (Figure 1A); as well as in patients with syndesmophyte formation/progression (n=18, 12 of them fulfilled also the mSASSS worsening criterion) versus those without (n=58): mean±SD 0.64±0.27, median: 0.65 μg/mL versus mean±SD 0.48±0.25, median 0.42 μg/mL, respectively, p=0.035 (Figure 1B).

A receiver operating characteristic (ROC)-analysis showed a good performance of calprotectin in prediction of mSASSS worsening (area under the curve [AUC]=0.740 (95% CI 0.614 to 0.866), p=0.004; Figure 2A), as well as in prediction of syndesmophyte formation/ progression (AUC=0.670 (95% CI 0.520 to 0.819), p=0.031; Figure 2B). In patients with axSpA who already had syndesmophytes at baseline (n=49; 65%), calprotectin demonstrated similar performance, compared to the whole group, as a predictor of radiographic spinal progression: AUC=0.741 (95% CI 0.597 to 0.884), p=0.009 for mSASSS worsening and AUC=0.706 (95% CI 0.547 to 0.866), p=0.020 for syndesmophyte formation/progression.

Calprotectin serum level >0.5 μg/mL (the optimal predictive cut-off in the ROC-analysis) had

108 CALPROTECTIN FOR RADIOGRAPHIC SPINAL PROGRESSION 7

Figure 1. Calprotectin serum levels in patients with and without radiographic spinal progression defined as an mSASSS worsening by ≥2 points over 2 years (A) and syndesmophyte formation/progression (B). Each box indicates the median value, the 1st and the 3rd quartiles; whiskers demonstrate minimal and maximal values, circles – outliers, crosses inside the boxes – mean values. P-values refer to the independent samples t-test. mSASSS, modified Stoke Ankylosing Spondylitis Spine Score.

Figure 2. Receiver Operating Characteristic (ROC) analysis of the association between baseline serum level of calprotectin and radiographic spinal progression defined as an mSASSS worsening by ≥2 points over 2 years (A) and syndesmophyte formation/progression (B) in patients with axial SpA (n=76).

CHAPTER 7 109 a sensitivity of 80%, a specificity of 62%, and an OR=6.2 (95% CI 1.6 to 24.2), p=0.009, as a predictor of mSASSS worsening. As a predictor of syndesmophyte formation/progression, calprotectin serum level >0.5 μg/mL had a sensitivity of 72%, a specificity of 60%, and an OR=4.0 (95% CI 1.2 to 12.6), p=0.020.

Although calprotectin serum level demonstrated a positive correlation with acute-phase reactants (CRP: Spearman’s r=0.382, p=0.001, ESR: r=0.401, p<0.001) and showed an overall performance similar to CRP in prediction of radiographic progression, the association of high calprotectin with radiographic spinal progression remained significant after adjustment for CRP and other risk factors (syndesmophytes at baseline and smoking): OR=5.5 (95% CI 1.2 to 25.8), p=0.030, for mSASSS worsening and an OR=4.1 (95% CI 1.2 to 32.0), p=0.028 for syndesmophyte formation/progression.

Also, calprotectin demonstrated a significant correlation with vascular endothelial growth factor (VEGF) – another biomarker with predictive value for radiographic spinal progression in axSpA:9 Spearman’s r=0.552, p<0.001. After adjustment for VEGF, calprotectin remained significantly associated with mSASSS worsening (OR=4.7 [95% CI 1.1 to 20.5], p=0.040) while the association with syndesmophyte formation has lost statistical significance (OR=3.6 [95% CI 1.0 to 13.7], p=0.056).

Thus, calprotectin represents a novel predictive biomarker for radiographic spinal progression in axial SpA.

110 CALPROTECTIN FOR RADIOGRAPHIC SPINAL PROGRESSION REFERENCES

1. Poddubnyy D, Haibel H, Listing J, et al. Baseline radiographic damage, elevated acute-phase reactant levels, and cigarette smoking status predict spinal radiographic progression in early axial spondylarthritis. Arthritis Rheum 2012;64:1388-98. 2. De Rycke L, Baeten D, Foell D, et al. Differential expression and response to anti-TNFalpha treatment of infiltrating versus resident tissue macrophage subsets in autoimmune arthritis. J Pathol 2005;206:17-27. 3. Kruithof E, De Rycke L, Vandooren B, et al. Identification of synovial biomarkers of response to experimental treatment in early-phase clinical trials in spondylarthritis. Arthritis Rheum 2006;54:1795-804. 4. Madland TM, Larsen A, Brun JG. S100 proteins calprotectin and S100A12 are related to radiographic changes rather than disease activity in psoriatic arthritis with low disease activity. J Rheumatol 2007;34:2089- 7 92. 5. Hammer HB, Odegard S, Syversen SW, et al. Calprotectin (a major S100 leucocyte protein) predicts 10-year radiographic progression in patients with rheumatoid arthritis. Ann Rheum Dis 2010;69:150-4. 6. Klingberg E, Carlsten H, Hilme E, et al. Calprotectin in ankylosing spondylitis--frequently elevated in feces, but normal in serum. Scand J Gastroenterol 2012;47:435-44. 7. Rudwaleit M, Haibel H, Baraliakos X, et al. The early disease stage in axial spondylarthritis: Results from the german spondyloarthritis inception cohort. Arthritis Rheum 2009;60:717-27. 8. Poddubnyy D, Rudwaleit M, Haibel H, et al. Effect of non-steroidal anti-inflammatory drugs on radiographic spinal progression in patients with axial spondyloarthritis: results from the German Spondyloarthritis Inception Cohort. Ann Rheum Dis 2012;71:1616-22. 9. Poddubnyy D, Conrad K, Haibel H, et al. Elevated serum level of the vascular endothelial growth factor predicts radiographic spinal progression in patients with axial spondyloarthritis. Ann Rheum Dis 2013 [Epub ahead of print].

CHAPTER 7 111

General discussion and summary 8 GENERAL DISCUSSION AND SUMMARY

Spondyloarthritis (SpA) is an inflammatory disease of the axial skeleton and peripheral joints which can be associated with extra-articular manifestations such as uveitis, psoriasis and inflammatory bowel disease (IBD). In this thesis we have addressed four key questions:

1) Can we find tools for early diagnosis? Currently, the delay between first symptoms and diagnosis is still 5-10 years. As non-steroidal anti-inflammatory drugs (NSAIDs) andanti- tumor necrosis factor (TNF) treatment are also effective in early disease, reducing this diagnostic and therapeutic delay could lead to major gains in overall health, function, and quality of life;

2) Can we use serum biomarkers to predict treatment response? Since a number of new treatments are available to treat SpA effectively, it is important to predict which patients will respond optimally to which treatment;

3) Can we identify outcome measures for peripheral SpA? We have well validated outcome measures in axial SpA which allow us to reliably determine the effectiveness of therapeutic interventions; this is currently lacking in peripheral SpA;

4) Can we find markers that predict structural damage? Structural damage in SpA is often characterized by new bone formation rather than by bone destruction (such as in RA). Although we can halt joint destruction with TNF blockade, we are currently not able to significantly modulate new bone formation.

This thesis contributes to addressing these challenges by exploring the value of serum biomarkers, either as single tools or in the context of composite outcome measures.

EARLY DIAGNOSIS

Current strategies The diagnostic delay of axial SpA is still approximately 5-10 years.1–3 Of note, data on the delay between first symptoms and diagnosis in peripheral SpA are still lacking. Diagnosing SpA in the early phase of the disease has become increasingly important, since 1) disease activity as measured by signs and symptoms of inflammation is similarly high in early and established disease,4,5 2) treatments used to treat signs and symptoms in established disease are equally effective in early disease,6–11 and 3) patients may respond better to treatment in the early phase of the disease.12 In the past decade, a number of strategies have been developed to capture the disease in the early phase.

114 GENERAL DISCUSSION AND SUMMARY The first is the development of early referral procedures from the general practitioner or orthopedic surgeons to the rheumatologists. Several referral criteria have been evaluated, using back pain, inflammatory back pain (IBP), human leukocyte antigen-B27 (HLA-B27) positivity, sacroiliitis on magnetic resonance imaging (MRI), positive family history, and good response to NSAIDs as most important criteria.13–17 Overall, the less complex strategies appeared to be most efficiently leading to the diagnosis of axial SpA by the rheumatologist in 16-45% of the referred patients.13–15,18 Still, the diagnostic delay remained above five years, indicating that there is still need for additional tools for the early detection and diagnosis of SpA.

The second strategy to capture axial SpA in the early phase is the use of MRI. In contrast to classical radiographic imaging, MRI can detect axial inflammation before the appearance of structural changes such as erosions and new bone formation. Although this is a very interesting topic to discuss, it is beyond the scope of this thesis.

To conclude, the combination of early referral strategies and the use of MRIs of the SI joints already decreased significantly the diagnostic delay of axial SpA, but the delay on average 8 still remains more than 5 years. The search for other strategies and tools to detect and identify the disease in (very) early phase thus remains important.

Peripheral blood biomarkers One potential way to decrease the diagnostic delay is the use of biomarkers. A biomarker is defined as a “characteristic that can be objectively measured and evaluated as an indicator of a normal biologic process, a pathophysiologic process, or a pharmacologic response to a therapeutic intervention”.23 Biomarkers can be derived of any tissue or body fluid. Focusing on one of the target tissues of SpA, the peripheral synovitis, our group has previously reported clear histological, cellular, and molecular differences between SpA and other rheumatic conditions.24–27 Although some of these features may also have a diagnostic value, sampling target tissues is a relatively complex and invasive procedure which cannot easily be used in daily clinical settings. Obviously, peripheral blood would be a much more convenient and reliable source for diagnostic biomarkers. Several peripheral blood-derived biomarkers have already been evaluated for diagnostic properties in SpA:

1) Genetic risk factors: the major genetic risk factor for ankylosing spondylitis (AS),the prototype of axial SpA, is HLA-B27. Although 80-90% of the AS patients carry the HLA-B27 gene (and 40-60% in other subtypes),28 HLA-B27 is also found in up to 8% of the general Western population. Accordingly, the positive predictive value of HLA-B27 is rather limited in the absence of a clear clinical suspicion.29,30 As discussed for MRI, testing for HLA-B27 positivity is also only meaningful in individuals with sufficient clinical suspicion of SpA.

2) Several serum markers, including C-reactive protein (CRP), erythrocyte sedimentation

CHAPTER 8 115 rate (ESR), and calprotectin, were shown to be elevated in full-blown active SpA compared to healthy controls, and to decrease upon effective treatment.31–33 In chapter 3, however, we demonstrated that these serum inflammatory biomarkers were not or, in case of calprotectin, only marginally elevated in the SPACE cohort (n=310) when comparing early axial SpA with ‘control’ early back pain patients, and that therefore these markers are not useful as early diagnostic markers. To screen for other new candidate biomarkers, we first performed a literature search in chapter 2 on informative biomarkers in related immune- mediated inflammatory diseases, namely psoriasis and IBD. We concluded that human beta defensin-2 (hBD-2)34–37 and lipocalin-2,38–44 were the most interesting candidates to test in SpA since these are interleukin (IL)-17 driven markers and since the IL-23/IL-17 is a major pathogenic cytokine pathway in SpA. Second, we identified IL-27, a cytokine belonging to the IL-23/IL-12 family, as an additional candidate maker since one recent report showed elevated serum levels in full-blown AS versus healthy controls.45 In chapter 3 we determined the serum levels of these three markers in AS and healthy controls but did not detect differences between the two groups. Therefore we did not test these markers in the SPACE cohort. We concluded that the aforementioned serum inflammatory biomarkers were not useful for diagnosing axial SpA at the early phase of the disease.

Two important aspects should be considered when interpreting these data. First, as we did not perform a systematic ‘omics’ approach to find diagnostic serum biomarkers but rather used a targeted approach based on literature data, it cannot be excluded that we have missed interesting, yet unknown serum diagnostic biomarkers. Second, our findings are in line with the concept that SpA is a tissue-specific disease rather than a systemic inflammatory disease and that therefore measuring markers in peripheral blood is not informative. This concept is further supported by our unpublished observations that a) disease-specific molecular and cellular signatures found in target tissues (e.g. myogene signature in SpA synovitis),27 cannot be detected in peripheral blood, and b) global transcriptomic analyses by pan-genomic microarray on peripheral blood cells of early SpA patients versus ‘control’ back pain patients from the SPACE cohort failed to reveal disease-specific expression signatures (Yeremenko and Turina, unpublished data).

3) Auto-antibodies: in contrast to many other rheumatic diseases, SpA is not characterized by the presence of disease-specific autoantibodies.46 Several approaches have been used to screen for potential autoantibodies but overall the results have been disappointing. Duftner et al.47 showed that serum concentrations of antibodies binding to the procaryotically expressed 28-kDa protein were higher in AS versus healthy controls. Chou et al.48 found that levels of anti-agalactosyl immunoglobulin-G antibody (anti-Gal(0) IgG) were elevated in AS and psoriatic arthritis when compared with healthy controls. In another study, plasma of AS, rheumatoid arthritis, and healthy controls, were screened for 3498 proteins, using two protein arrays to determine autoantibodies. In total, 44%, 33% and 8% of the individuals in the different groups had a broad autoantibody response, respectively.49 Nevertheless, all the

116 GENERAL DISCUSSION AND SUMMARY aforementioned results had low sensitivities and specificities, and were never reproduced in independent cohorts. Therefore these autoantibodies do not appear to be useful for diagnostic purposes.

Recently, one group demonstrated IgG-antibodies against CD74 specific for HLA class II- associated invariant chain peptide (CLIP) with high sensitivity (85.1%) and specificity (92.2%) for SpA.50,51 We are currently in the process of confirming these data in our cohorts. In contrast to the published data, and despite using the same assay in the same laboratory, IgG-antibodies against CD74 levels neither were elevated in full-blown AS versus healthy controls nor in early axial SpA versus early ‘control’ back pain patients. Nevertheless, ongoing analyses suggest that anti-CD74 IgA antibodies may be elevated in active AS versus healthy controls with a sensitivity of 38% and a specificity of 95%. In the near future, we will determine anti-CD74 IgA in the SPACE cohort to see whether these autoantibodies may have an early diagnostic value. Inception cohorts 8 A key requisite for the search for novel diagnostic biomarkers is the availability of appropriate inception cohorts. An inception cohort is a longitudinal cohort with a predefined group of persons gathered at a specific time point as early as possible in the development of a specific disorder. These cohorts could facilitate the search for valuable clinical, biological, and imaging biomarkers with the ultimate goal to dramatically decrease the time to diagnosis and start treatment without delay. In contrast to longitudinal studies, cross- sectional studies are found to be obsolete for this purpose.52 Several interesting inception cohort studies have been initiated over the last decade. GESPIC5 is an inception cohort of diagnosed AS and nr-axSpA patients with a maximum symptom duration of 10 and 5 years, respectively, which is very useful for the analysis of progression of structural damage. Esperanza53 is a cohort with patients having SpA features for at least 3 months and at most 2 years. DESIR54 and SPACE55 included patients with inflammatory back pain of 3 years at most, or back pain of 2 years at most, respectively. As described in chapter 4, we have initiated a different inception cohort study of seemingly healthy (i.e. individuals who are not known with musculoskeletal complaints by the general practitioner or rheumatologist) first-degree relatives (FDRs) of HLA-B27 positive patients (the so called ‘Pre-SpA’ cohort). The estimated recurrence risk for SpA in these relatives is 12%, which is much higher than the 0.5-1.5% risk in the general population.29,30,56 In contrast to other cohorts, in which clinical signs and symptoms (most often back pain) are required for inclusion, the Pre-SpA cohort thus focuses on baseline evaluation and prospective follow-up of non-symptomatic individuals at risk. In the first pilot analysis of Pre-SpA, we found that 17 out of 51 (33%) FDRs already fulfilled SpA classification criteria (the Assessment of Spondyloarthritis international Society [ASAS] axial SpA and/or the European Spondylarthropathy Study Group [ESSG] criteria)16,57 at baseline. Although fulfilling classification criteria does not necessarily imply the presence

CHAPTER 8 117 of a clinical diagnosis, these preliminary data suggest the presence of unrecognized disease in a significant proportion of FDRs. Interestingly, a similar proportion of HLA-B27+ and HLA-B27- FDRs fulfilled the classification criteria. Moreover, 6 of the 38 (16%) FDRs who did not fulfill the classification criteria had imaging abnormalities suggestive of SpA. Further follow-up of these patients will clarify who will develop established SpA, and will shed light on the question if these early abnormalities should be considered false positive. Inthe future, we will perform several other analyses in these FDRs (e.g. serum markers, genetics) in the search for new biomarkers with the ultimate goal to decrease the diagnostic delay of SpA. Based on the preliminary data from this pilot study, we are now coordinating a national study in 6 SpA centers of excellence (AMC, VUMC, LUMC, MUMC, UMCG, and Maasstad hospital) aimed to include and prospectively monitor 500 FDRs.

To conclude, diagnosing axial SpA at the early phase of the disease (at an earlier phase than currently possible) is an important but difficult goal. Early referral strategies and MRIs have resulted in a significant decrease of the diagnostic delay, but the delay should be further diminished. Investigations of serum biomarkers have not yet resulted in reliable diagnostic tools. Potential leads are the anti-CD74 IgA antibodies and the use of new ‘omics’ technologies. In all cases, the availability of large and well-documented early and preclinical SpA inception cohorts is essential for the evaluation and validation of diagnostic biomarkers.

PREDICTING TREATMENT RESPONSES

Treatment regimen Although the number of registered treatment for SpA is still limited when compared with rheumatoid arthritis, the therapeutic choices are rapidly expanding. In axial SpA, treatment options are limited to NSAIDs58 and TNF inhibitors.6,11,59,60 In peripheral SpA, patients can be treated with local steroids, disease-modifying antirheumatic drugs (DMARDs) such as sulfasalazine,61,62 and TNF inhibitors.9,10 In psoriatic arthritis, patients can be treated additionally with DMARDs such as leflunomide,63–65 apremilast66,67 and methotrexate,68,69 although the effectiveness of the latter DMARD is subject to debate despite common use in clinical practice. Ustekinumab (a monoclonal antibody against the p40 subunit of IL-23 and IL-12),70–72 is a non-TNF inhibitor biological recently registered for psoriatic arthritis. An overview of the currently available therapies is given in Table 1.

Besides the currently available drugs, an exciting new therapeutic option in both AS and psoriatic arthritis is blockade of IL-17, either by using monoclonal antibodies directed against IL-17A (such as secukinumab and ixekizumab) or by monoclonal antibodies blocking IL-17 receptors (such as brodalumab). Large phase III studies showed that secukinumab is effective and safe in both AS and psoriatic arthritis, and this drug is currently inthe

118 GENERAL DISCUSSION AND SUMMARY Table 1. An overview of currently available treatment options in ankylosing spondylitis and psoriatic arthritis Registered drug Generic name SpA subtype Mode of action Reference group NSAIDs Non-selective inhibitors: AS Prostaglandin inhibitor 58 aceclofenac, fenylbutazon, PsA ibuprofen, indomethacin, ketoprofen naproxen, piroxicam, etc. Selective COX-2 inhibitors: celecoxib, etoricoxib, meloxicam DMARDs/small Sulfasalazine AS Unknown 61,62 molecules (peripheral joint involvement) PsA Methotrexate PsA Folic acid antagonist 68,69 Leflunomide PsA Dihydro- 63–65 orotaatdehydrogenase inhibitor Apremilast PsA PDE-4 inhibitor 66,67 8 Biologicals TNF inhibitors: AS TNF inhibition 6,11,59,60 adalimumab, certolizumab- PsA 73–75 pegol, etanercept, golimumab, infliximab, Ustekinumab PsA IL-23 and IL-12 inhibitor 70–72 (p40 subunit) AS, ankylosing spondylitis; COX-2, cyclooxygenase-2; DMARDs, disease-modifying antirheumatic drugs; IL, interleukin; NSAIDs, non-steroidal anti-inflammatory drugs; PDE-4, Phosphodiesterase-4; PsA, psoriatic arthritis; SpA, spondyloarthritis; TNF, tumor necrosis factor.

registration phase.70,76,77 Many other drugs, including biologics as well as small molecules, are currently in phase II or III clinical trials in AS and/or psoriatic arthritis (Table 2), with many more in preclinical development. An extensive overview of recent and ongoing clinical trials in SpA was recently provided by Paramarta et al.78

The increasing number of treatment options urge an appropriate prediction, preferably prior to the start of therapy, about which patients will respond well to which treatments in the long-term. For example, 20-40% of the patients do not respond well to TNF inhibitors.56 In clinical practice, a significant part of these partial responders are continuing onTNF inhibitors since there are no other treatment options. With the advent of IL-17 blockade and hopefully other compounds in the near future, it becomes increasingly important to select an adequate strategy prior to the start of the treatment.

CHAPTER 8 119 Table 2. An overview of ongoing clinical trials in ankylosing spondylitis and psoriatic arthritis Drug group Clinical trial phase Mode of action SpA subtype Trial name and reference Small molecules Tofacitinib Phase 2/Phase 3 JAK inhibitor PsA OPAL

Thalidomide Phase 2 TNF and IL-12 Inhibitor AS NA Apremilast Phase 3 PDE-4 inhibitor AS POSTURE

Nilotinib Phase 2 Tyrosine kinase Peripheral Paramarta inhibitor and axial SpA et al, unpublished Biologicals Secukinumab Phase 3 RCT/ IL-17A inhibitor AS 79 Registration phase PsA 80 Ustekinumab Phase 2, open label IL-23 and IL-12 AS 81 trial inhibitor (p40 subunit) BI-655066 Phase 2 RCT, ongoing IL-23 inhibitor AS NA (p19 subunit) Brodalumab Phase 3 RCT IL-17 RA inhibitor PsA NA Ixekizumab Phase 3 RCT, ongoing IL-17A inhibitor PsA SPIRIT-P1 SPIRIT-P2 AS, ankylosing spondylitis; IL, interleukin; JAK, janus kinase; NA, not available; PDE-4, phosphodiesterase-4; PsA, psoriatic arthritis; RA, receptor A; RCT, randomized clinical trial.

Current strategies Several studies have focused on the identification of baseline predictors for treatment response in clinical trials at the group level in axial SpA. Of note, data on baseline predictors in peripheral SpA are lacking. Identifying such baseline predictors is important for designing new clinical studies, in which stratification at baseline is based on these predictors. A post- hoc analysis82 of two clinical trials with active AS patients (n=99) receiving infliximab versus placebo, or etanercept versus placebo and showed that age, Bath ankylosing spondylitis functional index (BASFI), disease duration, and elevated CRP/ESR levels were good predictors for treatment response at week 12, defined by Bath akylosing spondylitis disease activity index (BASDAI). Recently, another large pooled analysis was performed on four randomized controlled trials with etanercept versus sulfasalazine or placebo in 1281 active AS patients to determine the predictors for clinical remission as expressed by ASAS partial remission.83 In this analysis short disease duration of ≤ 2 years and age ≤ 40 years at the time of diagnosis, were good baseline predictors for clinical remission at week 12. Regarding biomarkers in this analysis, patients with elevated CRP levels showed better responses to treatment when compared with patients with normal baseline CRP levels. Similar results were found in other clinical trials with TNF inhibitors83–85 as well as in large observational cohorts with AS patients receiving TNF inhibitors.86,87 To summarize, the baseline predictors short disease

120 GENERAL DISCUSSION AND SUMMARY duration, young age at time of diagnosis, high BASFI scores, and elevated CRP/ESR levels are associated with good response to treatment at week 12. For the scope of this thesis, we have focused exclusively on the biomarkers.

Many recent clinical trials, and mainly those conducted in the early and non-radiographic forms of axSpA (nr-axSpA), have elevated CRP levels or a positive MRI of the sacroiliac (SI)-joints as an important eligibility criterion at screening and/or baseline, which seems justified based on the aforementioned results. Nevertheless, a sub-analysis of ABILITY-1,10 a phase 3 randomized controlled trial with nr-axSpA patients receiving adalimumab or placebo, showed that CRP is indeed a good predictor for clinical response at week 12, but not for response at year 1 and year 3. This finding suggests that CRP does reflect clinical response at the short term but does not reflect a ‘true responder’ who still responds well to adalimumab treatment at the long term. Moreover, CRP is a non-stable biomarker that can easily fluctuate between elevated and normal CRP levels over time, especially around the ‘upper limit of normal’ levels. For example, in the placebo arm of the same ABILITY-1 trial, 14 out of 57 (25%) patients with normal CRP levels at baseline converted into ‘elevated CRP’ between baseline and week 12.88 We recently confirmed the latter result in nr-axSpA 8 patients and AS patients, using the placebo arm of RAPID-axSpA, a phase-3 double-blind randomized controlled trial with certolizumab pegol (MSD) or placebo for 16 weeks.11 Thirteen of the 26 (50%) patients with normal CRP levels at baseline, had elevated levels of CRP in 1-6 consequent visits in the placebo arm (Turina et al., unpublished data). The results of these two analyses suggest that clinical trials should not use single measurements of CRP as a major eligibility criterion. Whether this is also true for MRI, is beyond the scope of this thesis.

The prediction of treatment response is not only important for stratification in clinical trials and for clinical research, but also for the decision-making process of the rheumatologist in individual patients in clinical practice. In AS for example, the clinical decision to start TNF inhibitors is based on a strategy that proclaims TNF inhibitor treatment could be started when a patient has consistently active disease, as expressed by BASDAI≥4, despite the optimal use of NSAIDs.89,90 Since the available effective treatments are expensive and come along with numerous side-effects, ideally the rheumatologist should be able to estimate in advance that a certain patient will have beneficial effects from the treatment. All the aforementioned parameters (including elevated CRP) are informative predictors at the group level, but fail to predict at the individual level.91–93 Thus, we need not only more reliable and robust markers of response to treatment but also markers with sufficient sensitivity and specificity to have a high predictive value in individual patients.

Measuring early changes after treatment

Since at the individual patient level biomarkers at baseline cannot predict whether patients will respond to a certain treatment, an interesting alternative concept might be to search for

CHAPTER 8 121 early changes of biomarker levels between baseline and shortly after the start of treatment (e.g. baseline and 2 weeks after treatment). One would expect significant decreases in certain biomarker levels over a short period of time when effective treatment is started. This theory is based on the finding that ‘true’ responders will respond early to treatment, as was published in a post-hoc analysis from the phase 3 ATLAS randomized controlled trial.94 Two hundred-two active AS patients of the total 311 (65%) participating patients received at least one dose of adalimumab and completed five years of follow-up. The best predictor for remission (as expressed by ASAS partial remission and Ankylosing Spondylitis Diseases Activity Score [ASDAS]-inactive disease at year 1 and year 5) was the achieved remission (by ASAS partial remission or ASDAS-inactive disease) after 12 weeks of treatment. Similar results were seen in the post-hoc analysis in peripheral SpA in the ABILITY-2 study, a phase 3 randomized controlled trial where patients received adalimumab or placebo at the first 12 weeks (double blind period) prior to the open label phase up to three years.95 ASDAS-inactive disease or Peripheral SpA Response Criteria (PSpARC) remission at week 12 were the best predictors for clinical remission (ASDAS-inactive disease or PSpARC remission) at years 1 and/or 3, when compared to baseline characteristics. The results of the two studies indicate that a ‘true’ responder to TNF inhibitors in the long term can be predicted by reaching an early response (short term outcome, i.e. 12 weeks of treatment) to this treatment rather than by baseline features alone.

An interesting approach to predict treatment responses very shortly after the startof treatment was demonstrated in chapter 5. Here we evaluated early changes in serum biomarker levels after TNF blockade was initiated. Serum was collected from axial and peripheral SpA patients of a randomized controlled trial, and two open label trials with TNF inhibitors. The latter two cohorts were considered as validation cohorts. These were all proof-of-concept trials, indicating that the sample sizes were relatively small. We selected seven serum biomarkers (hs-CRP, interleukin-6, pentraxin-3, alpha-2-macroglobulin, matrix metalloproteinase-3 [MMP-3], calprotectin, and vascular endothelial growth factor [VEGF]) and determined the levels at baseline and after two or four weeks of treatment, depending on the cohort design. We aimed to determine which of the biomarkers had a high sensitivity to rapid change upon effective treatment in SpA in proof-of-concept trials. We found that in all three cohorts, calprotectin and hs-CRP were good markers with high sensitivity to change upon effective treatment in both axial and peripheral SpA, whereas the levels didnot decrease in the placebo arm. We did not find similar results for the other five biomarkers. Although this finding was only based on TNF inhibitors, early changes of calprotectin and CRP levels were also detected in our proof-of-concept trial with secukinumab,76 thereby implying that early changes of calprotectin and CRP levels after effective treatment are not restricted to a certain mode of action. Although these results are only applicable at the group level in a proof-of-concept setting, the detection of early changes of biomarker levels is an interesting approach to further study at the individual level. Once this approach is applicable in clinical practice, the rheumatologist could decide on the treatment regimen

122 GENERAL DISCUSSION AND SUMMARY within two weeks, i.e. maintain the treatment or switch between TNF and IL-17 inhibitors in SpA.

IDENTIFYING OUTCOME MEASURES IN SPA

Objective markers for inflammation versus patient-reported outcomes Biomarkers may not only be useful for early diagnosis and prediction of treatment responses but also for monitoring disease activity in trials, clinical research as well as clinical practice. Historically, the BASDAI and the Visual Analogue Scales (VAS) for pain or patient global assessment of disease activity have been well validated and broadly used. These measures, however, have two major drawbacks.

First, as most outcome measures used in clinical research in the field of SpA, the BASDAI (a composite score of questions on tiredness, back pain, peripheral pain, enthesitis and 8 morning stiffness) and VAS measures rely completely on patient-reported outcomes (PROs) and therefore may be influenced by many other factors (e.g. intercurrent pathologies, psycho-social factors) besides genuine ‘biological disease activity’. The major issue here is that we still lack biomarkers that directly reflect disease activity in SpA, with CRP levels probably being currently the best ‘proxy’. Therefore, combination of PROs and objective measures such as CRP in a composite index might be the most informative manner to determine disease activity in SpA at the moment. One disease activity index that includes both PROs and an objective measure, is the Ankylosing Spondylitis Disease Activity Score (ASDAS).96,97 ASDAS is a composite of CRP levels and of questions on back pain, patient global, peripheral pain/swelling, and morning stiffness. The discriminatory ability of ASDAS is higher than that of BASDAI, a purely PRO driven composite index.96,98

Second, most disease activity measures have only been validated and used in axial disease, and it remains unknown if they also adequately reflect peripheral disease manifestations. Several other tools have been used to measure peripheral arthritis, enthesitis, and dactylitis in psoriatic arthritis, but these outcome measures then do not capture axial disease activity. As a large proportion of SpA patients display both axial and peripheral disease, we assessed in chapter 6 if composite disease activity indices used in axial SpA could also be used to assess peripheral SpA. Comparing several continuous outcome measures in two independent randomized controlled trials with adalimumab versus placebo in peripheral SpA, we showed that ASDAS and BASDAI performed equally well in terms of sensitivity to change and in discriminatory ability in peripheral disease. Since ASDAS does perform better than BASDAI in axial SpA, we recommend the use of ASDAS (which includes CRP) as preferred outcome measure in SpA.

CHAPTER 8 123 PREDICTING STRUCTURAL DAMAGE

In SpA, structural damage is characterized by bone and cartilage destruction as well as by new bone formation, which can eventually lead to total fusion (ankylosis) of the axial skeleton (the spine) and peripheral joints. Total fusion of the spine occurs in 28% and 43% of AS patients with a disease duration of 30-40 years and above 40 years, respectively.99 Accordingly, the structural damage due to new bone formation is an important co-determinant (together with active inflammation) of spinal mobility, function, and quality of life overtime.100 Whereas TNF blockers can potently inhibit inflammation and joint destruction in SpA, they unfortunately fail to significantly halt new bone formation. Therefore, an ultimate challenge is to halt or even prevent radiographic spinal progression. We will discuss here the different theories on new bone formation in SpA as well as potential strategies to predict which patients will suffer most from this aspect of the disease. Such prediction is very important for a number of reasons. First, it allows to reassure those patients who will not develop pronounced osteoproliferation. Second, it is needed to stratify patients in clinical research on structural damage in SpA. And third, it will be an essential determinant of individualized treatment if future research demonstrates that either very early treatment or drugs targeting different pathogenetic mechanisms (such as potentially the previously discussed IL-17 inhibitors) do halt new bone formation.

Inflammation and new bone formation There are three main theories about pathophysiological new bone formation. First, Sieper et al.101 hypothesized that remodeling is a repair process triggered by an initial phase of inflammation and tissue destruction. In case this theory is correct, TNF blockade might be able to halt new bone formation only when started in a very early disease stage, again emphasizing the importance of decreasing the diagnostic delay. The key question would then become which patients benefit most from such an early and aggressive treatment approach. Second, Lories et al.102 hypothesized that “entheseal stress” leads to triggering of two largely independent processes: an acute inflammatory reaction, which evolves then to chronic inflammation, and activation of progenitor cells that generate new bone. The uncoupling of inflammation and new bone formation in this model implies that, whereas all patients may require anti-inflammatory therapy, a specific treatment to target the bone remodelling pathway should be developed and given to patients prone to develop spinal new bone formation. Third, based on data in pre-clinical models, our group proposed that new bone formation is directly driven by inflammatory mediators distinct from soluble TNF, including transmembrane TNF (van Duivenvoorde et al., unpublished data) and IL-17A (van Tok et al., unpublished data). If this concept can be translated to human SpA, it would imply that it is crucial to identify which patient will not show extensive new bone formation and may thus be effectively treated with NSAID or TNF inhibitor versus those who may benefit

124 GENERAL DISCUSSION AND SUMMARY from other targeted treatments such as IL-17 blockade.

Current strategies in predicting new bone formation The “Outcome in AS International Study” (OASIS) cohort, a longitudinal study of 217 consecutive AS patients with a follow-up period of 12 years with clinical and imaging data, allowed to identify a series of clinical and imaging risk factors for radiographic progression of disease. The most important predictors in this study were high disease activity, smoking, HLA-B27 positivity, male gender, older age, and already existing syndesmophytes at baseline.103–105 Also serum biomarkers had been explored in the context of radiological progression of spinal disease, including CRP,106,107 ESR,103 matrix metalloproteinase (MMP- 3),108,109 dickkopf-1 (Dkk-1),110–112 visfatin, leptin,113–115 and VEGF.116,117 In chapter 7, we additionally determined the value of calprotectin using the GESPIC cohort. Calprotectin was found to be an independent predictor for new bone formation, but was not superior to CRP. It thus appears that several biomarkers are significantly associated with new bone formation at the group level but, unfortunately, none of these markers is sensitive and specific enough to be used for prediction of structural damage in individual patients. 8 Besides clinical features and biomarkers, multiple studies have explored the value of MRI visualization of inflammation and fatty changes as predictor of new bone formation. As the published results are partially discrepant118,119 and as imaging was not the focus of this thesis, we will not discuss this here in detail. However, it is crucial to realize that novel imaging technologies, including but not restricted to MRI, may form a very promising venue to measure objectively inflammatory and structural tissue changes in SpA and thereby may in the future be more appropriate than serum biomarkers for SpA detection, monitoring, and prediction. To illustrate this concept, we will briefly discuss here novel developments in the use of PET-CT in axial SpA.

PET-CT imaging with specific tracers for the detection of new bone formation Positron emission tomography-computed tomography (PET/CT) or PET/MRI allows to visualize the uptake of specific radiolabelled tracers in tissues and cells. For example, labelled glucose will be taken up by metabolically active tissues. Our colleagues form the VUMC recently explored the value of PET/CT with [18F]-fluoride as the uptake of this specific tracer in bone represents the blood flow and osteoblastic bone synthesis and activity at a specific location. A pilot study compared [18F]-fluoride PET/CT, MRI and X-ray of the spine in six TNF inhibitor naïve AS patients.120 An increased uptake of [18F]-fluoride on PET/CT was associated with corner inflammatory lesions (CIL) on MRI, and with existing syndesmophytes on the X-ray of the spine (OR=55.1; 95% CI, 7.3-422.3; p<0.001). Buchbender et al.121 studied PET/MRI-SI joint and spine in 13 active, TNF inhibitor naive AS patients according to the mNY criteria. They reported that [18F]-fluoride PET/MRI bone marrow edema and to a lesser extent fatty deposition was associated to osteoblastic activity, whereas structural changes (including

CHAPTER 8 125 sclerosis, erosions, ankylosis) were not frequently associated with osteoblastic activity. This is an interesting finding since the osteoblast activity was seen at sites where no structural changes were seen on MRI. Moreover, at several sites, elevated [18F]-fluoride uptake was seen whereas simultaneous MRI abnormalities were not depicted.

Although these studies are still preliminary and more data from follow-up studies are awaited, both studies illustrated well the potential of new molecular imaging techniques to visualize, measure, and monitor the disease processes in the target tissues of SpA. As our data presented here in this thesis tend to suggest that SpA disease processes are not easily captured by serum biomarkers and as most target tissues of SpA (such as spine, SI joints, and entheses) are difficult to approach, molecular imaging should probably be an area of major research investment in the future years.

126 GENERAL DISCUSSION AND SUMMARY REFERENCES

1. Sorensen J, Hetland ML. Decreases in diagnostic delay 84. are supported by sensitivity analyses. Ann Rheum Dis 14. Poddubnyy D, Vahldiek J, Spiller I, et al. Evaluation 2014;73(7):e45–e45. of 2 screening strategies for early identification of 2. Salvadorini, F. Bandinelli, A. Delle Sedie, L. Riente, A. patients with axial spondyloarthritis in primary care. J Candelieri, S. Generini, N. Possemato, S. Bombardieri Rheumatol 2011;38(11):2452–60. MM-C. Ankylosing spondylitis: how diagnostic and 15. Van Hoeven L, Vergouwe Y, de Buck PDM, Luime JJ, therapeutic delay have changed over the last six Hazes JMW, Weel AE a M. External Validation of a decades. Clin Exp Rheumatol 2012;30(4):561–5. Referral Rule for Axial Spondyloarthritis in Primary 3. Feldtkeller E, Khan MA, van der Heijde D, van der Care Patients with Chronic Low Back Pain. PLoS One Linden S, Braun J. Age at disease onset and diagnosis 2015;10(7):e0131963. delay in HLA-B27 negative vs. positive patients with 16. Sieper J, Rudwaleit M, Baraliakos X, et al. The ankylosing spondylitis. Rheumatol Int 2003;23(2):61–6. Assessment of SpondyloArthritis international Society 4. Boonen A, Sieper J, van der Heijde D, et al. The burden (ASAS) handbook: a guide to assess spondyloarthritis. of non-radiographic axial spondyloarthritis. Semin Ann Rheum Dis 2009;68 Suppl 2:ii1–44. Arthritis Rheum 2015;44(5):556–62. 17. Rudwaleit M, van der Heijde D, Landewé R, et al. The 5. Rudwaleit M, Haibel H, Baraliakos X, et al. The early Assessment of SpondyloArthritis International Society disease stage in axial spondylarthritis: results from the classification criteria for peripheral spondyloarthritis German Spondyloarthritis Inception Cohort. Arthritis and for spondyloarthritis in general. Ann Rheum Dis Rheum 2009;60(3):717–27. 2011;70(1):25–31. 6. Van Den Bosch F, Kruithof E, Baeten D, et al. Randomized 18. Sieper J, Srinivasan S, Zamani O, et al. Comparison double-blind comparison of chimeric monoclonal of two referral strategies for diagnosis of axial 8 antibody to tumor necrosis factor alpha (infliximab) spondyloarthritis: the Recognising and Diagnosing versus placebo in active spondylarthropathy. Arthritis Ankylosing Spondylitis Reliably (RADAR) study. Ann Rheum 2002;46(3):755–65. Rheum Dis 2013;72(10):1621–7. 7. Braun J, Brandt J, Listing J, et al. Treatment of 19. Rudwaleit M, van der Heijde D, Landewé R, et al. The active ankylosing spondylitis with infliximab: a development of Assessment of SpondyloArthritis randomised controlled multicentre trial. Lancet international Society classification criteria for axial 2002;359(9313):1187–93. spondyloarthritis (part II): validation and final selection. 8. Mease PJ, Goffe BS, Metz J, VanderStoep A, Finck Ann Rheum Dis 2009;68(6):777–83. B, Burge DJ. Etanercept in the treatment of psoriatic 20. De Rycke L, Maas M, Tak PP, Baeten D. “MRI-tis” in the arthritis and psoriasis: a randomised trial. Lancet early diagnosis of axial SpA: issues and limitations. Nat 2000;356(9227):385–90. Rev Rheumatol 2010;6(11):666–9. 9. Paramarta JE, De Rycke L, Heijda TF, et al. Efficacy and 21. Van der Heijde D, Rudwaleit M, Landewe R, Sieper J. safety of adalimumab for the treatment of peripheral Justification for including MRI as a tool in the diagnosis arthritis in spondyloarthritis patients without of axial SpA. Nat Rev Rheumatol 2010;6:670–2. ankylosing spondylitis or psoriatic arthritis.Ann Rheum 22. Berg R van den, Hooge M de, Navarro-Compán V, et Dis 2013;72(11):1793–9. al. How useful is imaging of the SI-joint (MRI and/ 10. Sieper J, van der Heijde D, Dougados M, et al. Efficacy or X-ray) in patients with possible spondyloarthritis and safety of adalimumab in patients with non- in the diagnostic work-up? Ann Rheum Dis radiographic axial spondyloarthritis: results of a 2012;71(Suppl3):566 randomised placebo-controlled trial (ABILITY-1). Ann 23. Gruttola VG De, Clax P, Demets DL, et al. Considerations Rheum Dis 2013;72(6):815–22. in the Evaluation of Surrogate Endpoints in Clinical 11. Landewé R, Braun J, Deodhar a, et al. Efficacy of Trials: Summary of a National Institutes of Health certolizumab pegol on signs and symptoms of axial Workshop. Control Clin Trials 2001;1998(2001):485– spondyloarthritis including ankylosing spondylitis: 502. 24-week results of a double-blind randomised 24. Baeten D, Kruithof E, Van den Bosch F, et placebo-controlled Phase 3 study. Ann Rheum Dis al. Immunomodulatory effects of anti-tumor 2014;73(1):39–47. necrosis factor alpha therapy on synovium in 12. Seo MR, Baek HL, Yoon HH, et al. Delayed diagnosis is spondylarthropathy: histologic findings in eight linked to worse outcomes and unfavourable treatment patients from an open-label pilot study. Arthritis responses in patients with axial spondyloarthritis. Clin Rheum 2001;44(1):186–95. Rheumatol 2014; 25. Baeten D, Kruithof E, De Rycke L, et al. Infiltration of 13. Brandt HC, Spiller I, Song I-H, Vahldiek JL, Rudwaleit the synovial membrane with macrophage subsets M, Sieper J. Performance of referral recommendations and polymorphonuclear cells reflects global disease in patients with chronic back pain and suspected axial activity in spondyloarthropathy. Arthritis Res Ther spondyloarthritis. Ann Rheum Dis 2007;66(11):1479– 2005;7(2):R359–69.

CHAPTER 8 127 26. Noordenbos T, Yeremenko N, Gofita I, et al. Interleukin- and disease activity. J Gastroenterol 2012;47(5):519– 17-positive mast cells contribute to synovial 30. inflammation in spondylarthritis. Arthritis Rheum 40. Yeşil A, Gönen C, Senateş E, et al. Relationship between 2012;64(1):99–109. neutrophil gelatinase-associated lipocalin (NGAL) levels 27. Yeremenko N, Noordenbos T, Cantaert T, et al. Disease- and inflammatory bowel disease type and activity. Dig specific and inflammation-independent stromal Dis Sci 2013;58(9):2587–93. alterations in spondylarthritis synovitis. Arthritis 41. Ataseven A, Kesli R, Kurtipek GS, Ozturk P. Assessment Rheum 2013;65(1):174–85. of lipocalin 2, clusterin, soluble tumor necrosis factor 28. Thomas GP, Brown M a. Genetics and genomics of receptor-1, interleukin-6, homocysteine, and uric ankylosing spondylitis. Immunol Rev 2010;233(1):162– acid levels in patients with psoriasis. Dis Markers 80. 2014;2014:541709. 29. Akkoc N. Are spondyloarthropathies as common 42. El-Hadidi H, Samir N, Shaker OG, Otb S. Estimation of as rheumatoid arthritis worldwide? A review. Curr tissue and serum lipocalin-2 in psoriasis vulgaris and Rheumatol Rep 2008;10(5):371–8. its relation to metabolic syndrome. Arch Dermatol Res 30. Reveille JD, Witter JP, Weisman MH. Prevalence of axial 2014;306(3):239–45. spondylarthritis in the United States: estimates from 43. Kamata M, Tada Y, Tatsuta a, et al. Serum lipocalin-2 a cross-sectional survey. Arthritis Care Res (Hoboken) levels are increased in patients with psoriasis. Clin Exp 2012;64(6):905–10. Dermatol 2012;37(3):296–9. 31. Benhamou M, Gossec L, Dougados M. Clinical relevance 44. Romaní J, Caixàs A, Ceperuelo-Mallafré V, et al. of C-reactive protein in ankylosing spondylitis and Circulating levels of lipocalin-2 and retinol-binding evaluation of the NSAIDs/coxibs’ treatment effect protein-4 are increased in psoriatic patients and on C-reactive protein. Rheumatology (Oxford) correlated with baseline PASI. Arch Dermatol Res 2010;49(3):536–41. 2013;305(2):105–12. 32. Visvanathan S, Wagner C, Marini JC, et al. Inflammatory 45. Lin T-T, Lu J, Qi C-Y, et al. Elevated serum level of IL-27 biomarkers, disease activity and spinal disease and VEGF in patients with ankylosing spondylitis and measures in patients with ankylosing spondylitis associate with disease activity. Clin Exp Med 2014; after treatment with infliximab. Ann Rheum Dis 46. Ambarus C, Yeremenko N, Tak PP, Baeten D. 2008;67(4):511–7. Pathogenesis of spondyloarthritis: autoimmune 33. De Vries MK, van Eijk IC, van der Horst-Bruinsma or autoinflammatory? Curr Opin Rheumatol IE, et al. Erythrocyte sedimentation rate, C-reactive 2012;24(4):351–8. protein level, and serum amyloid a protein for patient 47. Duftner C, Dejaco C, Klauser a, Falkenbach a, Lakomek selection and monitoring of anti-tumor necrosis factor HJ, Schirmer M. High positive predictive value of treatment in ankylosing spondylitis. Arthritis Rheum specific antibodies cross-reacting with a 28-kDa 2009;61(11):1484–90. Drosophila antigen for diagnosis of ankylosing 34. Kim JM. Antimicrobial Proteins in Intestine spondylitis. Rheumatology (Oxford) 2006;45(1):38–42. and Inflammatory Bowel Diseases. Intest Res 48. Chou C-L, Wu M-J, Yu C-L, et al. Anti-agalactosyl 2014;9100(1):20–33. IgG antibody in ankylosing spondylitis and psoriatic 35. Yamaguchi N, Isomoto H, Mukae H, et al. arthritis. Clin Rheumatol 2010;29(8):875–81. Concentrations of alpha- and beta-defensins in plasma 49. Wright C, Sibani S, Trudgian D, et al. Detection of of patients with inflammatory bowel disease. Inflamm multiple autoantibodies in patients with ankylosing Res 2009;58(4):192–7. spondylitis using nucleic acid programmable protein 36. Jansen P a M, Rodijk-Olthuis D, Hollox EJ, et al. arrays. Mol Cell Proteomics 2012;11(2):M9.00384. Beta-defensin-2 protein is a serum biomarker for 50. Baraliakos X, Baerlecken N, Witte T, Heldmann F, Braun disease activity in psoriasis and reaches biologically J. High prevalence of anti-CD74 antibodies specific relevant concentrations in lesional skin. PLoS One for the HLA class II-associated invariant chain peptide 2009;4(3):e4725. (CLIP) in patients with axial spondyloarthritis. Ann 37. Gambichler T, Bechara FG, Scola N, Rotterdam S, Rheum Dis 2014;73(6):1079–82. Altmeyer P, Skrygan M. Serum levels of antimicrobial 51. Baerlecken NT, Nothdorft S, Stummvoll GH, et al. peptides and proteins do not correlate with psoriasis Autoantibodies against CD74 in spondyloarthritis. Ann severity and are increased after treatment with fumaric Rheum Dis 2014;73(6):1211–4. acid esters. Arch Dermatol Res 2012;304(6):471–4. 52. Kiltz U, Baraliakos X, Karakostas P, et al. The degree 38. Janas RM, Ochocińska A, Snitko R, et al. Neutrophil of spinal inflammation is similar in patients with gelatinase-associated lipocalin in blood in children with axial spondyloarthritis who report high or low levels inflammatory bowel disease. J Gastroenterol Hepatol of disease activity: a cohort study. Ann Rheum Dis 2014;29(11):1883–9. 2012;71(7):1207–11. 39. Oikonomou KA, Kapsoritakis AN, Theodoridou C, et 53. Muñoz-Fernández S, Carmona L, Collantes E, et al. A al. Neutrophil gelatinase-associated lipocalin (NGAL) model for the development and implementation of in inflammatory bowel disease: association with a national plan for the optimal management of early pathophysiology of inflammation, established markers,

128 GENERAL DISCUSSION AND SUMMARY spondyloarthritis: the Esperanza Program. Ann Rheum 67. Kavanaugh A, Mease PJ, Gomez-Reino JJ, et Dis 2011;70(5):827–30. al. Treatment of psoriatic arthritis in a phase 3 54. Dougados M, d’Agostino M-A, Benessiano J, et randomised, placebo-controlled trial with apremilast, al. The DESIR cohort: a 10-year follow-up of early an oral phosphodiesterase 4 inhibitor. Ann Rheum Dis inflammatory back pain in France: study design and 2014;73(6):1020–6. baseline characteristics of the 708 recruited patients. 68. Kingsley GH, Kowalczyk A, Taylor H, et al. A randomized Joint Bone Spine 2011;78(6):598–603. placebo-controlled trial of methotrexate in psoriatic 55. Van den Berg R, de Hooge M, van Gaalen F, Reijnierse arthritis. Rheumatology (Oxford) 2012;51(8):1368–77. M, Huizinga T, van der Heijde D. Percentage of patients 69. Soriano ER, McHugh NJ. Therapies for peripheral joint with spondyloarthritis in patients referred because of disease in psoriatic arthritis. A systematic review. J chronic back pain and performance of classification Rheumatol 2006;33(7):1422–30. criteria: experience from the Spondyloarthritis 70. McInnes IB, Sieper J, Braun J, et al. Efficacy and safety Caught Early (SPACE) cohort. Rheumatology (Oxford) of secukinumab, a fully human anti-interleukin-17A 2013;52(8):1492–9. monoclonal antibody, in patients with moderate- 56. Dougados M, Baeten D. Spondyloarthritis. Lancet to-severe psoriatic arthritis: a 24-week, randomised, 2011;377(9783):2127–37. double-blind, placebo-controlled, phase II proof-of- 57. Dougados M, Linden S Van Der, Juhlin R, et al. The concept trial. Ann Rheum Dis 2014;73(2):349–56. European Spondylarthropathy Study Group Preliminary 71. Ritchlin C, Rahman P, Kavanaugh A, et al. Efficacy and Criteria for the Classification of Spondylarthropathy. safety of the anti-IL-12/23 p40 monoclonal antibody, Arthritis Rheum 1991;34(10):1218–27. ustekinumab, in patients with active psoriatic arthritis 58. Zochling J, van der Heijde D, Dougados M, Braun J. despite conventional non-biological and biological Current evidence for the management of ankylosing anti-tumour necrosis factor therapy: 6-month and spondylitis: a systematic literature review for the ASAS/ 1-year results of the phase 3, multicentre, double-blind, EULAR management recommendations in ankylosing placebo-controlled PSUMMIT 1 trial. Ann Rheum Dis 8 spondylitis. Ann Rheum Dis 2006;65(4):423–32. 2014;73(6):990–9. 59. Brandt J, Khariouzov a, Listing J, et al. Six-month 72. Gottlieb A, Menter A, Mendelsohn A, et al. Ustekinumab, results of a double-blind, placebo-controlled trial of a human interleukin 12/23 monoclonal antibody, for etanercept treatment in patients with active ankylosing psoriatic arthritis: randomised, double-blind, placebo- spondylitis. Arthritis Rheum 2003;48(6):1667–75. controlled, crossover trial. Lancet (London, England) 2009;373(9664):633–40. 60. Van der Heijde D, Kivitz A, Schiff MH, et al. Efficacy and safety of adalimumab in patients with ankylosing 73. Mease PJ, Gladman DD, Ritchlin CT, et al. Adalimumab spondylitis: results of a multicenter, randomized, for the treatment of patients with moderately to double-blind, placebo-controlled trial. Arthritis Rheum severely active psoriatic arthritis: Results of a double- 2006;54(7):2136–46. blind, randomized, placebo-controlled trial. Arthritis Rheum 2005;52(10):3279–89. 61. Chen J, Liu C. Is sulfasalazine effective in ankylosing spondylitis? A systematic review of randomized 74. Mease PJ, Kivitz AJ, Burch FX, et al. Etanercept controlled trials. J Rheumatol 2006;33(4):722–31. treatment of psoriatic arthritis: Safety, efficacy, and effect on disease progression. Arthritis Rheum 62. Clegg DO, Reda DJ, Abdellatif M. Comparison of 2004;50(7):2264–72. sulfasalazine and placebo for the treatment of axial and peripheral articular manifestations of the 75. Antoni CE, Kavanaugh A, Kirkham B, et al. Sustained seronegative spondylarthropathies: A Department of benefits of infliximab therapy for dermatologic Veterans Affairs Cooperative Study. Arthritis Rheum and articular manifestations of psoriatic arthritis: 1999;42(11):2325–9. Results from the Infliximab Multinational Psoriatic Arthritis Controlled Trial (IMPACT). Arthritis Rheum 63. Kaltwasser JP, Nash P, Gladman D, et al. Efficacy and 2005;52(4):1227–36. safety of leflunomide in the treatment of psoriatic arthritis and psoriasis: A multinational, double-blind, 76. Baeten D, Baraliakos X, Braun J, et al. Anti-interleukin- randomized, placebo-controlled clinical trial. Arthritis 17A monoclonal antibody secukinumab in treatment Rheum 2004;50(6):1939–50. of ankylosing spondylitis: a randomised, double-blind, placebo-controlled trial. Lancet 2013;382(9906):1705– 64. Liang GC, Barr WG. Open trial of leflunomide for 13. refractory psoriasis and psoriatic arthritis. J Clin Rheumatol 2001;7(6):366–70. 77. McInnes IB, Mease PJ, Kirkham B, et al. Secukinumab, a human anti-interleukin-17A monoclonal antibody, 65. Behrens F, Finkenwirth C, Pavelka K, et al. Leflunomide in patients with psoriatic arthritis (FUTURE 2):a in psoriatic arthritis: results from a large European randomised, double-blind, placebo-controlled, phase 3 Arthritis Care Res prospective observational study. trial. Lancet 2015;386(9999):1137-46. 2013;65(3):464–70. 78. Paramarta JE, Baeten D. Spondyloarthritis: 66. Kavanaugh a., Mease PJ, Gomez-Reino JJ, et al. from unifying concepts to improved treatment. Longterm (52-week) Results of a Phase III Randomized, Rheumatology (Oxford) 2014;53(9):1547–59. Controlled Trial of Apremilast in Patients with Psoriatic Arthritis. J Rheumatol 2015;42(3). 79. Sieper J, Braun J, Baraliakos X, et al. Secukinumab

CHAPTER 8 129 Significantly Improves Signs and Symptoms of Active 90. Vander Cruyssen B, Ribbens C, Boonen A, et Ankylosing Spondylitis: 52-Week Data from Measure 2, al. The epidemiology of ankylosing spondylitis A Randomized, Double-Blind, Placebo-Controlled Phase and the commencement of anti-TNF therapy 3 Trial with Subcutaneous Loading and Maintenance in daily rheumatology practice. Ann Rheum Dis Dosing. Ann Rheum Dis 2015;74(suppl2):132.3–133. 2007;66(8):1072–7. 80. Kavanaugh A., McInnes IB, Hall S, et al. Secukinumab 91. Davis J, van der Heijde D, Dougados M, et al. Baseline Efficacy in Anti-TNF-Naive and Anti-TNF-IR Patients Factors Influential on ASAS 20 Response in Ankylosing with Psoriatic Arthritis: Results of a Phase 3 Multicenter, Spondylitis Patients Treated With Etanercept Double-Blind, Placebo-Controlled Study (Future 2). Ann (ENBREL®). Arthritis Rheum 2003;48(Suppl.):S440. Rheum Dis 2015;74(suppl2):345.3–346. 92. Burgos-Vargas R, Rojas-Serrano J. Predictors of 81. Poddubnyy D, Hermann K-G a, Callhoff J, Listing J, response to tumor necrosis factor-alpha blockers in Sieper J. Ustekinumab for the treatment of patients ankylosing spondylitis. J Rheumatol 2005;32(9):1637– with active ankylosing spondylitis: results of a28- 40. week, prospective, open-label, proof-of-concept study 93. Arends S, van der Veer E, Kallenberg CGM, Brouwer (TOPAS). Ann Rheum Dis 2014;73(5):817–23. E, Spoorenberg A. Baseline predictors of response to 82. Rudwaleit M, Listing J, Brandt J, Braun J, Sieper J. TNF-α blocking therapy in ankylosing spondylitis. Curr Prediction of a major clinical response (BASDAI 50) to Opin Rheumatol 2012;24(3):290–8. tumour necrosis factor alpha blockers in ankylosing 94. Sieper J, van der Heijde D, Dougados M, Brown LS, Lavie spondylitis. Ann Rheum Dis 2004;63(6):665–70. F, Pangan AL. Early response to adalimumab predicts 83. Baraliakos X, Koenig AS, Jones H, Szumski A, Collier long-term remission through 5 years of treatment in D, Bananis E. Predictors of Clinical Remission under patients with ankylosing spondylitis. Ann Rheum Dis Anti-tumor Necrosis Factor Treatment in Patients with 2012;71(5):700–6. Ankylosing Spondylitis: Pooled Analysis from Large 95. Van den Bosch F, Mease P, Sieper J, et al. Early clinical Randomized Clinical Trials. J Rheumatol 2015;42(8): response is a better predictor of long-term remission 1418-26. than baseline disease characteristics following 84. Vastesaeger N, van der Heijde D, Inman RD, et al. adalimumab treatment in peripheral spondyloarthritis. Predicting the outcome of ankylosing spondylitis Ann Rheum Dis 2015;74(Suppl2):283. therapy. Ann Rheum Dis 2011;70(6):973–81. 96. Lukas C, Landewé R, Sieper J, et al. Development of 85. Rudwaleit M, Claudepierre P, Wordsworth P, et al. an ASAS-endorsed disease activity score (ASDAS) in Effectiveness, safety, and predictors of good clinical patients with ankylosing spondylitis. Ann Rheum Dis response in 1250 patients treated with adalimumab 2009;68(1):18–24. for active ankylosing spondylitis. J Rheumatol 97. Machado P, Landewé R, Lie E, et al. Ankylosing 2009;36(4):801–8. Spondylitis Disease Activity Score (ASDAS): defining cut- 86. Arends S, Brouwer E, van der Veer E, et al. Baseline off values for disease activity states and improvement predictors of response and discontinuation of tumor scores. Ann Rheum Dis 2011;70(1):47–53. necrosis factor-alpha blocking therapy in ankylosing 98. Garrett S, Jenkinson T, Kennedy LG, Whitelock H, spondylitis: a prospective longitudinal observational Gaisford P, Calin A. A new approach to defining disease cohort study. Arthritis Res Ther 2011;13(3):R94. status in ankylosing spondylitis: the Bath Ankylosing 87. Glintborg B, Ostergaard M, Krogh N, Dreyer L, Spondylitis Disease Activity Index. Rheumatology Kristensen H, Hetland M. Predictors of treatment 1994;21(12):2286–91. response and drug continuation in 842 patients with 99. Jang JH, Ward MM, Rucker AN, et al. Ankylosing ankylosing spondylitis treated with anti-tumour Spondylitis : Patterns of Radiographic Involvement — A necrosis factor: results from 8 years’ surveillance in the Re-examination of Accepted Principles in a Cohort of Danish nationwide DANBIO registry. Ann Rheum Dis 769 Patients. Radiology 2011;258(1):192–8. 2010;69(11):2002–8. 100. Machado P, Landewe R, Braun J, Hermann K-G a., 88. Baraliakos X, Sieper J, Chen S, Pangan AL, Anderson J. Baker D, van der Heijde D. Both structural damage and Clinically active non-radiographic axial spondyloarthritis inflammation of the spine contribute to impairment of patients initially have a negative MRI and normal CRP spinal mobility in patients with ankylosing spondylitis. may develop a positive MRI or elevated CRP at a later Ann Rheum Dis 2010;69(8):1465–70. time point. In: Clinically active non-radiographic axial spondyloarthritis patients initially have a negative 101. Sieper J, Appel H, Braun J, Rudwaleit M. Critical MRI and normal CRP may develop a positive MRI or appraisal of assessment of structural damage in elevated CRP at a later time point. Ann Rheum Dis ankylosing spondylitis: implications for treatment 2015;74(Suppl2):83. outcomes. Arthritis Rheum 2008;58(3):649–56. 89. Braun J, Davis J, Dougados M, Sieper J, van der Linden S, 102. Lories RJU, Luyten FP, de Vlam K. Progress in van der Heijde D. First update of the international ASAS spondylarthritis. Mechanisms of new bone formation consensus statement for the use of anti-TNF agents in in spondyloarthritis. Arthritis Res Ther 2009;11(2):221. patients with ankylosing spondylitis. Ann Rheum Dis 103. Van Tubergen A, Ramiro S, van der Heijde D, Dougados 2006;65(3):316–20. M, Mielants H, Landewé R. Development of new syndesmophytes and bridges in ankylosing spondylitis

130 GENERAL DISCUSSION AND SUMMARY and their predictors: a longitudinal study. Ann Rheum radiographic spinal progression in patients with axial Dis 2012;71(4):518–23. spondyloarthritis. Ann Rheum Dis 2014;73(12):2137– 104. Ramiro S, van der Heijde D, van Tubergen A, et al. 43. Higher disease activity leads to more structural 117. Baraliakos X, Hermann KG, Xu S, Hsu B, Braun J. VEGF damage in the spine in ankylosing spondylitis: 12-year and CRP serum levels lack predictive value for outcomes longitudinal data from the OASIS cohort. Ann Rheum as assessed by conventional radiographs and magnetic Dis 2014;73(8):1455–61. resonance imaging in patients with active ankylosing 105. Ramiro S, Stolwijk C, van Tubergen A, et al. Evolution spondylitis treated with the TNF inhibitor golimumab. of radiographic damage in ankylosing spondylitis: a Ann Rheum Dis 2014;73(Suppl2):1138-1139. 12 year prospective follow-up of the OASIS study. Ann 118. Van der Heijde D, Machado P, Braun J, et al. MRI Rheum Dis 2015;74(1):52–9. inflammation at the vertebral unit only marginally 106. Poddubnyy D, Rudwaleit M, Haibel H, et al. Rates and predicts new syndesmophyte formation: a multilevel predictors of radiographic sacroiliitis progression over analysis in patients with ankylosing spondylitis. Ann 2 years in patients with axial spondyloarthritis. Ann Rheum Dis 2012;71(3):369–73. Rheum Dis 2011;70(8):1369–74. 119. Maksymowych WP, Chiowchanwisawakit P, Clare 107. Pedersen SJ, Sørensen IJ, Lambert RGW, et al. T, Pedersen SJ, Østergaard M, Lambert RGW. Radiographic progression is associated with resolution Inflammatory lesions of the spine on magnetic of systemic inflammation in patients with axial resonance imaging predict the development of new spondylarthritis treated with tumor necrosis factor α syndesmophytes in ankylosing spondylitis evidence of inhibitors. Arthritis Rheum 2011;63(12):3789–800. a relationship between inflammation and new bone formation. Arthritis Rheum 2009;60(1):93–102. 108. Yang C, Gu J, Rihl M, et al. Serum levels of matrix metalloproteinase 3 and macrophage colony- 120. Fischer DR, Pfirrmann CWA, Zubler V, et al. High stimulating factor 1 correlate with disease bone turnover assessed by 18F-fluoride PET / CT activity in ankylosing spondylitis. Arthritis Rheum in the spine and sacroiliac joints of patients with 8 2004;51(5):691–9. ankylosing spondylitis: comparison with inflammatory lesions detected by whole body MRI. Ann Rheum Dis 109. Arends S, van der Veer E, Groen H, et al. Serum MMP- 2012;71(Suppl3):606. 3 level as a biomarker for monitoring and predicting response to etanercept treatment in ankylosing 121. Buchbender C, Ostendorf B, Ruhlmann V, et al. spondylitis. J Rheumatol 2011;38(8):1644–50. Hybrid 18F-labeled Fluoride Positron Emission Tomography/Magnetic Resonance (MR) Imaging of the 110. Heiland GR, Appel H, Poddubnyy D, et al. High level Sacroiliac Joints and the Spine in Patients with Axial of functional dickkopf-1 predicts protection from Spondyloarthritis: A Pilot Study Exploring the Link of syndesmophyte formation in patients with ankylosing MR Bone Pathologies and Increased Osteoblastic Ac. Ann Rheum Dis spondylitis. 2012;71(4):572–4. J Rheumatol First Release July 1 2015; doi:10.3899/ 111. Klingberg E, Nurkkala M, Carlsten H, Forsblad-d’Elia jrheum.150250. H. Biomarkers of bone metabolism in ankylosing spondylitis in relation to osteoproliferation and osteoporosis. J Rheumatol 2014;41(7):1349–56. 112. Korkosz M, Jerzy G, Leszczy P, Pawlak-bu K, Jeka S, Kucharska E. High disease activity in ankylosing spondylitis is associated with increased serum sclerostin level and decreased wingless protein-3a signaling but is not linked with greater structural damage. BMC Musculoskelet Disord 2013;14:99. doi:10.1186/1471-2474-14-99. 113. Syrbe U, Callhoff J, Conrad K, et al. Serum adipokine levels in patients with ankylosing spondylitis and their relationship to clinical parameters and radiographic spinal progression. Arthritis Rheumatol (Hoboken, NJ) 2015;67(3):678–85. 114. Genre F, López-Mejías R, Miranda-Filloy J a, et al. Adipokines, biomarkers of endothelial activation, and metabolic syndrome in patients with ankylosing spondylitis. Biomed Res Int 2014;2014:860651. 115. Kim K-J, Kim J-Y, Park S-J, et al. Serum leptin levels are associated with the presence of syndesmophytes in male patients with ankylosing spondylitis. Clin Rheumatol 2012;31(8):1231–8. 116. Poddubnyy D, Conrad K, Haibel H, et al. Elevated serum level of the vascular endothelial growth factor predicts

CHAPTER 8 131

APPENDICES

Nederlandse samenvatting Dankwoord PhD portfolio Curriculum vitae List of publications & 134 APPENDICES NEDERLANDSE SAMENVATTING

Spondyloartritis (SpA) is een frequent voorkomende vorm van chronische gewrichtsontsteking die zowel het axiale skelet (sacro-iliacaal gewrichten en de wervelkolom) als de perifere gewrichten (met name oligoartritis [1-5 gewrichten] van de grote gewrichten zoals knieën en enkels) aantast. Axiale SpA kenmerkt zich door ontstekingsachtige rugpijn. Deze rugpijn uit zich door pijn in rust, verbetering bij beweging, nachtelijke pijn en ochtendstijfheid. Het proces van spinale ontsteking verloopt gelijktijdig met pathologische botnieuwvorming, die op zijn beurt kan leiden tot totale verbening van de sacro-iliacaal (SI-)gewrichten en de wervelkolom. Het prototype van axiale SpA is ‘spondylitis ankylopoietica’, ook wel de ziekte van Bechterew genoemd. Perifere SpA symptomen bestaan uit pijnlijke, warme, stijve en gezwollen gewrichten, en uit enthesitis (ontsteking ter hoogte van een peesaanhechting) en/of dactylitis (ontsteking van een gehele vinger [‘worstvinger’] of teen [‘worstteen’]). Bij deze vorm van SpA kan zowel botdestructie als pathologische botnieuwvorming worden gezien. Het prototype van perifere SpA is artritis psoriatica. Daarnaast kunnen extra- articulaire manifestaties voorkomen zoals ontstekingen van de ogen (uveitis anterior), de & darmen (inflammatoire darmziekten [IBD]: de ziekte van Crohn en colitis ulcerosa) en van de huid (psoriasis).

De behandeling van SpA in zijn algemeenheid is momenteel nog tamelijk beperkt: patiënten met axiale SpA worden behandeld met ‘non-steroidal antinflammatory drugs’ (NSAIDs) en (bij onvoldoende effect) met tumor necrosis factor (TNF)-alpha-remmers. Patiënten met perifere SpA kunnen worden behandeld met NSAIDs, maar ook met ‘disease-modifying antirheumatic drugs’ (DMARDs) en (bij onvoldoende effect) met TNF-remmers. Bij beide vormen van SpA hebben de huidige behandelingen goed effect op de symptomen en remmen zij bij perifere SpA ook botdestructie. Bij axiale SpA wordt het proces van botnieuwvorming echter niet geremd door de huidige medicatie.

Dit proefschrift In dit proefschrift trachten wij vier belangrijke vragen te beantwoorden over de diagnose, behandeling en de uitkomst van SpA: 1) diagnose: kunnen we (bio-)markers identificeren die bijdragen aan het diagnosticeren van SpA in de vroege (klinische en subklinische) fase van de ziekte? Het is belangrijk de periode tussen het ontstaan van symptomen en het diagnosticeren (de zogenaamde ‘diagnostic delay’) van axiale SpA, die gemiddeld 5tot 10 jaar duurt, te verkorten. 2) Respons op behandeling: kunnen we serum biomarkers identificeren voor het voorspellen van respons op nieuwe therapieën in kleinschalige, kortdurende ‘proof-of-concept’ (PoC) klinische trials? 3) Respons op behandeling: kunnen we uitkomstmaten valideren voor het voorspellen van respons op therapie in perifere SpA? Deze vragen zijn van belang vanwege de beperkte therapeutische mogelijkheden bij SpA. 4) Prognose: kunnen we biomarkers identificeren die voorspellend zijn voor de progressie van

135 structurele schade in de wervelkolom in axiale SpA? De pathologische botnieuwvorming in axiale SpA komt slechts bij een minderheid van de patiënten voor en het lijkt belangrijk deze groep van tevoren te kunnen identificeren. In dit proefschrift brengen wij de klinische en biologische markers in kaart om te trachten deze vragen te beantwoorden.

HET VROEGTIJDIG STELLEN VAN EEN DIAGNOSE SPA

Het vroegtijdig stellen van een diagnose is van belang omdat dan een tijdige behandeling kan worden gestart en omdat bekend is dat de behandeling in een late fase van de ziekte de botnieuwvorming niet remt. Er zijn enkele inflammatoire biomarkers in het serum bekend die mogelijk informatief zijn met betrekking tot SpA, namelijk c-reactief proteïne (CRP), bezinking (BSE) en calprotectine, een heterodimeer bestaande uit S100A8 en A9. In hoofdstuk 3 hebben wij deze drie markers getest in een cohort van patiënten met rugpijn met een duur tussen de 3 maanden en 2 jaar, maar ontstaan is voor het 45e levensjaar (SPondyloarthritis Caught Early [SPACE] cohort). Er werden echter geen belangrijke verschillen aangetoond tussen vroege axiale SpA patiënten (n=119) en ‘controle’ rugpijn patiënten (n=191). Hieruit kan worden geconcludeerd dat deze markers niet zinvol zijn voor vroegdiagnostiek in SpA. Om deze reden hebben we in hoofdstuk 2 bestudeerd welke biomarkers mogelijk bijdragend zijn bij aan SpA gerelateerde ziekten, namelijk psoriasis en IBD. Uit deze review blijkt dat humaan bèta defensin-2 (hBD-2) en lipocaline-2 bijdragend zouden kunnen zijn, vooral vanuit een pathofysiologisch oogpunt. HBD-2 en lipocaline-2 behoren tot de groep van antimicrobiële peptiden en zijn interessant bij SpA omdat beide peptiden worden gestuurd door interleukine (IL)-17. Uit eerder onderzoek in SpA is gebleken dat het cytokine IL-17 (vanuit de IL-23/IL-17 as) een belangrijke rol speelt bij het ontstaan van SpA. In hoofdstuk 3 hebben wij vervolgens onderzocht of er verschillen zijn in de hBD- 2 en lipocaline-2 serumwaardes tussen spondylitis ankylopoetica patiënten en gezonde individuen. Tevens hebben we bekeken of er verschillen zijn in de serumwaardes van IL-27 tussen deze twee groepen, aangezien dit was gerapporteerd in één andere studie. Geen van de drie biomarkers bleek te zijn verhoogd in ons cohort van spondylitis ankylopoetica patiënten in vergelijking met de gezonde individuen. Vanwege deze negatieve resultaten hebben we de experimenten niet meer herhaald in het SPACE cohort met vroege axiale SpA en controle patiënten.

Concluderend kunnen we stellen dat we met gebruik van de bovengenoemde biomarkers voor ontsteking een diagnose van SpA niet kunnen vervroegen.

136 APPENDICES Inceptie-cohorten De inceptie-cohorten vormen een belangrijk onderdeel in het identificeren van klinische, radiologische en biologische markers om SpA in een vroege fase te kunnen opsporen. Een inceptie-cohort is een cohort bestaande uit een groep van individuen met één ofmeer kenmerken die worden verzameld en vervolgd, bijvoorbeeld om te onderzoeken of zij een bepaalde ziekte ontwikkelen. Wij hebben in hoofdstuk 4 een inceptie-cohort ‘Pre- SpA’ opgezet met ‘op het eerste gezicht gezonde’ eerstegraads familieleden van HLA-B27 positieve spondylitis ankylopoetica patiënten (dit betekent: geen gediagnosticeerde SpA en geen klachten van het bewegingsapparaat waarvoor ze onder behandeling zijn bij huisarts of reumatoloog). Deze groep is gekozen omdat bekend was dat zij een verhoogd risico (12% versus 0,5-1,5% in de gewone populatie) hebben op het ontwikkelen van SpA. De familieleden tussen de 18-40 jaar zullen 5 jaar worden opgevolgd om te evalueren wie uiteindelijk SpA zal ontwikkelen. Deze leeftijdscategorie is gekozen omdat SpA zich voornamelijk manifesteert tussen het 20 en 40e levensjaar. Bij een analyse van de baseline visites van de eerste 51 deelnemers bleek dat 17 (33%) familieleden al konden worden geclassificeerd als hebbende SpA volgens de ASAS axiale SpA en/of de ‘European Spondylarthropathy Study Group‘ & (ESSG) criteria. Alhoewel niet alle geclassificeerde deelnemers daadwerkelijk een klinische diagnose zullen hebben (of krijgen), lijkt een aanzienlijk deel van de gezonde eerstegraads familieleden van SpA patiënten toch al klinische manifestaties van SpA hebben die in dit stadium echter nog geen aanleiding waren geweest om een arts te bezoeken, dan wel niet waren herkend horend bij SpA in de praktijk.

Vooralsnog zijn er in elk geval geen biologische markers (CRP, BSE, calprotectine) gevonden die zo’n diagnose kunnen voorspellen. Van de 38 eerstegraads familieleden die niet met SpA geclassificeerd werden, bleek er bij 6 (16%) familieleden tóch sprake te zijn van ontstekingslaesies op de MRI van de SI-gewrichten. Inclusie van meer deelnemers en het langdurig opvolgen van de familieleden zal uitwijzen wie uiteindelijk klinisch manifest SpA zal ontwikkelen. Bovendien kan er dan onderzocht worden of bepaalde markers voorspellend zijn voor het ontwikkelen van SpA.

Het ‘Pre-SpA’ cohort heeft nu reeds een substantieel aantal individuen kunnen classificeren die SpA in de vroege fase van de ziekte zouden kunnen hebben. Na een langere follow-up zal blijken welke klinische, radiologische en biologische markers zouden kunnen bijdragen aan het verkorten van de ‘diagnostic delay’.

HET VOORSPELLEN VAN DE RESPONS OP BEHANDELING

De verschijnselen van SpA kunnen bij een deel van de patiënten goed worden onderdrukt met de huidige medicatie, namelijk NSAIDs en TNF remmers. Maar er zijn een aantal

137 beperkingen. Allereerst reageert 40% van de patiënten onvoldoende goed op de ingestelde therapie. Ten tweede kunnen sommige patiënten niet starten in verband met bestaande contra-indicaties of moeten zij na de start alsnog (snel) de behandeling staken in verband met bijwerkingen. Ten derde wordt het proces van botnieuwvorming niet geremd. En ten slotte kan remissie op lange termijn vaak niet worden bereikt: de ziekte vlamt meestal weer op binnen enkele maanden na het staken van TNF blokkade. Anders geformuleerd: de huidige therapieën zijn effectief, maar mede gezien bovengenoemde nadelen, is het belangrijk om nieuwe behandelingen te ontwikkelen. Het is medisch-ethisch niet verantwoord om gedurende langere tijd en op grote schaal patiënten aan klinische trials te laten deelnemen met een gerede kans op een niet-werkende (experimentele) behandeling (of placebo) terwijl er al bewezen effectieve behandelingen bestaan. Een mogelijke oplossing hiervoor is de strategie met de PoC trials, waarbij relatief kort na de start van de behandeling kan worden onderscheiden of een nieuwe behandeling mogelijk effectief zal zijn, dan wel waarschijnlijk ineffectief. PoC trials zijn klinische trials van korte duur met een kleine groepsgrootte. Bij een gunstige uitkomst kunnen grotere fase 2b/fase 3 klinische trials worden gestart voor verdere bewijsvoering.

In hoofdstuk 5 hebben we gekeken naar vroege veranderingen in serum biomarkers vóór en na de start van TNF remmers bij SpA patiënten. Hiervoor zijn een aantal biomarkers geselecteerd, namelijk hoog-sensitief (hs-)CRP, interleukine-6, pentraxine-3, alpha-2- macroglobuline, matrix metalloproteinase-3 [MMP-3], calprotectine en vascular endothelial growth factor [VEGF]). De biomarker serumwaardes zijn gemeten in 3 onafhankelijke SpA cohorten ten tijde van baseline (t=0) en kort na start van behandeling (t=2 of t=4 weken). Gekeken werd bijvoorbeeld of er snelle dalingen (in 2-4 weken tijd) te zien waren in de bloedwaardes in de effectieve groep die niet optraden in de placebogroep. Uit deze analyses bleek dat alleen calprotectine en hs-CRP veranderden tijdens het gebruik van TNF remmers en stabiel bleven in de placebogroep. Vergelijkbare resultaten werden onlangs gevonden bij een nieuw effectief medicament, namelijk secukinumab dat zorgt voor IL-17 blokkade. Dit geeft aan dat deze twee biomarkers niet enkel van potentieel nut kunnen zijn bij het monitoren van het effect van TNF remmers maar mogelijk ook bij medicamenten met een andere ‘mechanism of action’ toepasbaar zijn. De gevoeligheid voor verandering van deze twee (objectieve) biomarkers bleek vergelijkbaar te zijn met die van door de patiënt gerapporteerde uitkomstmaten ‘Bath ankylosing spondylitis disease activity index’ (BASDAI) en de ‘ziekteactiviteit volgens de patiënt’.

Concluderend hebben de calprotectine en hs-CRP biomarkers een hoge gevoeligheid voor verandering bij effectieve therapie op groepsniveau in kleinschalige, kortdurende PoC trials.

138 APPENDICES HET IDENTIFICEREN VAN UITKOMSTMATEN

In axiale SpA zijn er diverse uitkomstmaten en dichotome responscriteria ontwikkeld en gevalideerd om het effect van een therapie te kunnen evalueren. In perifere SpA zijn deze maten niet gevalideerd, met uitzondering van artritis psoriatica, waarbij de ‘American College of Rheumatology’ (ACR) en ‘psoriatic arthritis response criteria’ (PsARC) worden toegepast. Een responsmaat ontwikkeld voor perifere SpA is de set van ‘peripheral spondyloarthritis response criteria’ (pSpARC), maar deze index was nog niet uitgebreid gevalideerd. In hoofdstuk 6 hebben we een post-hoc analyse uitgevoerd in twee placebo-gecontroleerde, gerandomiseerde klinische trials met perifere SpA patiënten (n=165 en n=40), waarbij de ene helft van de patiënten de TNF remmer adalimumab ontvingen en de andere helft placebo. Diverse uitkomstmaten werden met elkaar vergeleken. De resultaten tonen dat de scores ‘ziekteactiviteit volgens patiënt´ (patient global) en ´ziekteactiviteit volgens de dokter´ (physician global), globale uitkomstmaten voor meerdere ziektebeelden, alsook de BASDAI en de ‘Ankylosing spondylitis disease activity score’ (ASDAS) indices (gevalideerd voor spondylitis ankylopoietica en toegepast bij axiale SpA), de best discriminerende & uitkomstmaten zijn voor klinische trials in perifere SpA. De beste presterende responscriteria zijn de PsPARC en de ACR.

Concluderend hebben de ‘ziekteactiviteit volgens patiënt’ en ‘volgens dokter’, de BASDAI en de ASDAS goede discriminatoire eigenschappen voor het gebruik in klinische trials bij patiënten met perifere SpA. Omdat deze uitkomstmaten natuurlijk in eerste instantie zijn ontwikkeld voor axiale SpA (probleem met ‘face validity’) dienen in de toekomst indices te worden ontwikkeld die meer gericht zijn op symptomen en verschijnselen van perifere SpA. Van de geteste dichotome uitkomstmaten zijn PSpARC en ACR het beste in het discrimineren tussen een succesvolle en niet succesvolle behandeling.

HET VOORSPELLEN VAN STRUCTURELE SCHADE

Structurele schade bij SpA wordt enerzijds gekenmerkt door botdestructie en anderzijds door botnieuwvorming. Botdestructie kan goed worden geremd met TNF blokkade bij perifere SpA (artritis psoriatica). Maar TNF blokkade is niet geschikt om botnieuwvorming bij axiale SpA te remmen. Botnieuwvorming bij axiale SpA kan leiden tot volledige ankylosering van de wervellichamen. Ankylosering leidt tot vermindering van de mobiliteit en functie van de rug en vermindering van kwaliteit van leven. Het voorspellen wie botnieuwvorming en ankylose ontwikkelt, wordt van belang zodra dit proces kan worden afgeremd. Met de huidige therapieën is het nog niet mogelijk het proces af te remmen, maar er zijn aanwijzingen dat bij behandeling in een zeer vroeg ziektestadium dit wellicht wel mogelijk is. Ook zijn er veel therapieën in onderzoek en ontwikkeling, zoals IL-17 blokkade, die mogelijk

139 de botnieuwvorming kunnen remmen. Het is dus in toenemende mate van belang inzicht te krijgen in wie de meeste kans heeft op het ontwikkelen van botnieuwvorming opdat adequate behandeling kan worden gestart bij de juiste patiënten (prognostiek). Betere prognostiek leidt ook tot adequate stratificatie voor toekomstige klinische trials gericht op het remmen van botnieuwvorming. De resultaten van hoofdstuk 7 tonen dat serum calprotectine axiale progressie op röntgenfoto’s van de lumbale en cervicale wervelkolom, een radiologische maat voor botnieuwvorming, kan voorspellen. Calprotectine is echter niet beter dan de veelvuldig gebruikte maat CRP, waarvan in een eerder stadium al de voorspellende waarde werd aangetoond.

Concluderend hebben wij in dit proefschrift getracht bij te dragen aan de verbetering van vroegdiagnostiek van SpA en aan het voorspellen van een behandelingsrespons en het optreden van botnieuwvorming. We hebben aangetoond dat onze selectie van inflammatoire biomarkers in serum niet bijdragen aan de vroegdiagnostiek, maar enkele ervan mogelijk wel informatief zijn voor het signaleren van effectieve therapieën in PoC trials en voor het voorspellen van botnieuwvorming bij axiale SpA patiënten. Ten tweede hebben we uitkomstmaten gevalideerd bij perifere SpA. En ten slotte hebben wij het ‘Pre- SpA’ inceptie-cohort opgezet met eerstegraads familieleden van spondylitis ankylopoetica patiënten, waarbij in een aanzienlijk deel van de patiënten SpA kon worden geclassificeerd in een vroeg stadium.

140 APPENDICES &

141 PHD PORTFOLIO

Name PhD student M.C. Turina

PhD period September 2010 - August 2015 General courses The Academic Medical Center (AMC) World of Science 2010 0.7 ECTS Basiscursus Regelgeving en Organisatie Klinisch onderzoekers (BROK) AMC 2011 0.9 ECTS Practical Biostatistics AMC 2011 1.1 ECTS Scientific Writing in English for Publication AMC 2012 1.5 ECTS Clinical Epidemiology AMC 2013 0.6 ECTS Career Development AMC 2015 0.6 ECTS

Specific courses Federation of Clinical Immunology Societies (FOCIS) Advance Course in 2011 1.0 ECTS Basic & Clinical Immunology, Scottsdale, USA American College of Rheumatology (ACR) Basic research pre-conference: 2011 0.5 ECTS bone pathophysiology in inflammatory and rheumatic disorders, Chicago, USA & School of Translational Immunology, the European network for Translational 2012 0.7 ECTS Immunology Research and Education (ENTIRE), Belgrade, Serbia Postgraduate course Advanced Immunology AMC/VUmc/Sanquin 2012 2.9 ECTS Epidemiology Course of European League Against Rheumatism (EULAR), 2013 0.5 ECTS Berlin, Germany

Seminars and workshops Course Snel lezen Tijdwinst.com 2011 0.2 ECTS Course Time Management Tijdwinst.com 2011 0.2 ECTS Course Mind mapping Tijdwinst.com 2011 2.2 ECTS Course Presence and presentation, MGI coaching 2011 0.2 ECTS Weekly department research seminars AMC 2010-2015 12 ECTS Weekly department clinical education AMC 2010-2013 8 ECTS Lecturing Spondyloarthritis preceptorship for Novartis 2012 0.5 ECTS Rheumatology introduction course for third year medical students 2012 0.2 ECTS

Presentations Serum biomarkers to predict clinical response in proof-of-concept trials in spondyloarthritis: European workshop on Immune-Mediated Inflammatory Diseases (ewIMID), 2011 0.5 ECTS Nice, France (poster presentation) Translational Research Day Institut Pasteur, Paris, France (poster 2011 0.5 ECTS presentation) ACR/Association of Reproductive Health professionals (ARHP), Annual 2011 0.5 ECTS Scientific Meeting, Chicago, USA (poster presentation) Efficacy and safety of adalimumab for the treatment of peripheral arthritis in spondyloarthritis patients without ankylosing spondylitis or psoriatic arthritis:

147 Nederlandse vereniging voor reumatologie (NVR) Najaarsdagen, Papendal, 2011 0.5 ECTS the Netherlands (oral presentation) Baseline elevated serum levels of calprotectin as independent marker for radiographic spinal progression in ankylosing spondylitis: European Workshop for Rheumatology Research (EWRR), Prague, Czech 2012 0.5 ECTS (poster tour presentation) A psychometric analysis of outcome measures in trials of peripheral spondyloarthritis: EULAR Annual European Congress of Rheumatology, Paris, France (poster 2014 0.5 ECTS presentation) International Congress on Spondyloarthritis, Ghent, Belgium (poster 2014 0.5 ECTS presentation) Calprotectin (S100A8/A9) as serum biomarker for clinical response in proof- of-concept trials in axial and peripheral spondyloarthritis: EULAR Annual European Congress of Rheumatology, Paris, France (oral 2014 0.5 ECTS presentation) International Congress on Spondyloarthritis, Ghent, Belgium (poster 2014 0.5 ECTS presentation) Clinical and imaging signs of spondyloarthritis in first-degree relatives of HLA-B27 positive ankylosing spondylitis patients: the pre-spondyloarthritis (Pre-SpA) cohort: EULAR Annual European Congress of Rheumatology, Rome, Italy (poster 2015 0.5 ECTS tour presentation)

Conferences ewIMID, Sitges, Spain 2010 0.75 ECTS ewIMID, Nice, France 2011 0.75 ECTS Translational Research Day Institut Pasteur, Paris 2011 0.25 ECTS ACR/ARHP Annual Scientific Meeting, Chicago, USA 2011 1.5 ECTS NVR Najaarsdagen, Papendal, the Netherlands 2012 0.25 ECTS EWRR Workshop, Prague, Czech 2013 0.75 ECTS EULAR Annual European Congress of Rheumatology, Madrid, Spain 2013 1.0 ECTS EULAR Annual European Congress of Rheumatology, Paris, France 2014 1.0 ECTS International congress on Spondyloarthritis, Ghent, Belgium 2014 0.75 ECTS EULAR Annual European Congress of Rheumatology, Rome, Italy 2015 1.0 ECTS

148 APPENDICES &

149 LIST OF PUBLICATIONS

1 Turina MC*, Ramiro S*, Baeten DL, Mease P, Paramarta JE, Song I-H, Pangan AL, Landewé R. A psychometric analysis of outcome measures in peripheral spondyloarthritis.Ann Rheum Dis 2015;1–6. doi:10.1136/annrheumdis-2014-207235 *equal contributions

2 Vogelpoel LTC, Hansen IS, Rispens T, Muller FJM, van Capel TMM, Turina MC, Vos JB, Baeten DLP, Kapsenberg ML, de Jong EC, den Dunnen J. Fc gamma receptor-TLR cross- talk elicits pro-inflammatory cytokine production by human M2 macrophages. Nat Commun 2014;5:5444. doi:10.1038/ncomms6444

3 Turina MC, Yeremenko N, Paramarta JE, De Rycke L, Baeten D. Calprotectin (S100A8/9) as serum biomarker for clinical response in proof-of-concept trials in axial and peripheral spondyloarthritis. Arthritis Res Ther 2014;16:413. doi:10.1186/s13075- 014-0413-4 & 4 Turina MC, Sieper J, Yeremenko N, Conrad K, Haibel H, Rudwaleit M, Baeten D, Poddubnyy D. Calprotectin serum level is an independent marker for radiographic spinal progression in axial spondyloarthritis. Ann Rheum Dis 2014;73:1746–8. doi:10.1136/annrheumdis-2014-205506

5 Popadic D, Anegon I, Baeten D, Eibel H, Giese T, Marits P, Martinez-Caceres E, Mascart F, Nestle F, Pujol-Borrell R, Savic E, Scheibenbogen C, Seliger B, Thunberg S, Turina M, Villanova F, Winqvist O, Wikström A-C. Predictive immunomonitoring — The COST ENTIRE initiative. Clin Immunol 2013;147:23–6. doi:10.1016/j.clim.2013.01.013

153