sign in sign up
<<
Home , Ankylosing spondylitis

THE EVOLVING TREATMENT OF ANKYLOSING VRIJE UNIVERSITEIT

THE EVOLVING TREATMENT OF ANKYLOSING SPONDYLITIS

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus prof.dr. F.A. van der Duyn Schouten, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de Faculteit der Geneeskunde op woensdag 6 mei 2015 om 11.45 uur in de aula van de universiteit, De Boelelaan 1105

ISBN: 978-94-6259-628-3 Copyright © 2015 JC van Denderen Cover: Ellen van Diek Layout: Persoonlijkproefschrift.nl, Matthijs Ariëns Printing: Ispkamp Drukkers, Enschede door

Printing of thesis was financially supported by: BV, Janssen-Cilag BV, AbbVie Johannes Christiaan van Denderen BV, Mundipharma BV, UCB Pharma BV, Will-Pharma BV.

geboren te Kampen 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. promotor: prof.dr. J.W.J. Bijlsma CONTENTS copromotoren: dr. I.E. van der Horst-Bruinsma dr. M.T. Nurmohamed Chapter 1 Introduction: 1. General introduction. 8 2. History of the disease ankylosing spondylitis. 11 3. History of in ankylosing spondylitis. 21 4. Objectives and outline of this thesis. 29 Chapter 2 Efficacy and safety of mesalazine (Salofalk®) in an open study 41 of 20 patients with ankylosing spondylitis. Chapter 3 Statin therapy might be beneficial for treating patients with 51 ankylosing spondylitis. Chapter 4 Double blind, randomised, placebo controlled study of 57 in the treatment of active ankylosing spondylitis. Chapter 5 Decreased clinical response to in ankylosing 71 spondylitis is correlated with anti-infliximab formation. Chapter 6 Elevated liver enzymes in patients with ankylosing 81 spondylitis treated with . Chapter 7 What do we miss? ASAS non-responders on anti-TNF 95 therapy show improvement in performance-based physical function. Chapter 8 significantly reduces the recurrence rate of 107 anterior in patients with ankylosing spondylitis. Chapter 9 Etanercept increases mineral density in ankylosing 121 spondylitis, but does not prevent the increase of vertebral fractures. Chapter 10 1. General discussion and summary. 138 2. Nederlandse samenvatting. 162 3. Dankwoord. 168 4. Curriculum vitae. 169 5. List of publications. 170 6. List of abbreviations and illustrations. 173 CHAPTER 1

Introduction

1. General introduction. 2. History of the disease ankylosing spondylitis. 3. History of therapy in ankylosing spondylitis. 4. Objectives and outline of this thesis. Chapter 1 Introduction

1 GENERAL INTRODUCTION. limitation of mobility of the spine and thorax. After many years flattening of the lumbar and increase of the thoracic can occur; which produce The subject of this thesis is the disease ankylosing spondylitis (AS) and particularly the characteristic curved and immobile posture. In addition, this posture becomes various aspects of drug therapy in AS. aggravated by flexion in the hips in case of hip . Some AS AS is the prototype of a group of diseases, called spondyloarthritides, which have patients can not stand erect, nor see well straight forward (figure 1). The severity of 1 of the spine and sacroiliac as common features. Other diseases AS can range from mild disease and limited radiological changes to a severe disease belonging to spondyloarthritis (SpA) are , , arthritis with persisting disease activity and disability. Pain and structural damage can associated with inflammatory bowel disease and a subgroup of juvenile idiopathic cause limitation of mobility, functional impairment, reduction in well being and can arthritis. In the Netherlands AS is often called “ziekte van Bechterew”. lead to absence and disability at work [8]. Predictors of functional impairment are: 2 AS is characterised by a chronic sterile inflammation of joints and entheses worse initial features of the disease, older age, smoking, uveitis and peripheral (entheses are the sites of attachment of tendons, ligaments and joint capsules). disease, especially hip involvement [6, 9-12]. The inflammation in AS is localised primarily in the sacroiliac joints and spine, Patients with AS have an increased risk of mortality [10]. In a review the standardized 3 but peripheral joints can be involved as well. The peripheral arthritis is usually an mortality ratio was approximately 1.7 (range 1.5-1.9), which for the most part could asymmetric of the lower extremities, including the hips. In contrast to be ascribed to cardiovascular diseases [13]. the inflammation does in a lesser degree lead to destruction (erosions) of bone, but can lead to local ossification. This hyperostosis can lead In AS extra-spinal and extra-articular manifestations are observed. AS is associated 4 to and in some cases to the characteristic bamboo spine. This has given with skin disease (), inflammatory bowel disease ( and the disease its name: the Greek ankylos means stiff/curved and spondylos means Crohn’s disease) and preceding enteral or genitor-urinal infections (reactive /spine. arthritis). One third of the patients with AS experience attacks of inflammation of the eye (acute anterior uveitis) [14]. Problems of the (conduction disturbance 5 In mid-Europe AS affects 0.3-0.5 % of the population and the prevalence for SpA is and aortic root anomaly) and sometimes apical do occur. 1-2% [1]. The (HLA) B27 is the most important genetic The incidence of and vertebral fractures is increased, as well as the factor linked to the disease. In more than 90% of Caucasian AS patients HLA-B27 is incidence of cardiovascular disease [15-18]. 6 present. This explains the familial aggregation of AS and the higher prevalence of AS in the northern countries, where HLA-B27 is more prevalent [2]. Men are affected In the next paragraph, the historical context of the disease AS will be discussed, more often than women (ratio 2-3 to 1) [1, 3]. including the evolution towards the recognition as a separate disease. In the third The pathophysiology of the disease is not fully clarified. Probably, in subjects with paragraph of the introduction the therapy of AS in the past and present will be 7 genetic susceptibility, auto-inflammatory processes are triggered by environmental discussed. At the end of the introduction an outline of this thesis is given. factors (for example enteral or urogenital microbes and smoking), mechanical stress and/or HLA-B27 binding peptides [4-6]. 8 Laboratory investigation can show raised acute phase reactants (erythrocyte sedimentation rate (ESR) and C-reactive protein), but often does not [7]. Diagnostic for AS are radiological changes of the sacroiliac joints: sclerosis, erosions and ankylosis. Later in the disease process similar changes can be seen in the rest of the 9 spine at the corners of the spinal corpora, in the intervertebral and costovertebral joints. 10 In most cases the disease starts in young adults (<30 years), but is diagnosed several years later. General symptoms, like anaemia and tiredness, often occur. The inflammation causes pain and stiffness in the buttocks, the lower back and other parts of the spine. The complaints are worst in the late night and early morning and improve with moving and not with rest. Inflammation and ankylosis can cause

8 9 Chapter 1 Introduction

2 HISTORY OF THE DISEASE ANKYLOSING SPONDYLITIS.

The evolution of ankylosing spondylitis (AS) towards a separate disease entity is described in this paragraph, based on old descriptions, pathophysiological findings in animal and human skeletons and new scientific developments like the 1 invention of X-rays and DNA-sequencing, that unravelled the HLA-B27 association. Classification criteria for AS were developed and later AS was embedded in the broader spectrum of (SpA). 2 Old literature and art Luke, one of the four evangelists, was a Greek who wrote around Anno Domini 75 in verse 13 of the Gospel ascribed to him: “Ende siet / daer was eene 3 vrouwe / die hadde eenen geest der kranckheyt achtien jaren / ende sy was krom / ende en konde niet wel op-sien” [19]. Caelius Aurelianus, a Roman physician in the fifth century, described a patient in Figure 1. Progression of ankylosing spondylitis in a patient with severe disease. By 1957, 10 “De Ischiadicis et Psoadicis”: The patient is seized by pain in the nates, moves years after the onset of disease, he has very limited extension of his spine as evidenced by the 4 loss of lordosis and exaggeration of thoracic kyphosis (spine is fusing in flexion). By 1967, he slowly, and can only bend or stand erect with difficulty [20]. is unable to extend his cervical spine. He developed contractures at the knees and hip disease In old literature and visual arts, people with increased kyphosis were described (note the cane) leading to in 1973 with subsequent improvement of posture. or depicted and some very interesting publications have covered this subject [21- (ACR Slide Collection) 23]. To some of these cases the diagnosis ankylosing spondylitis (AS) could apply, 5 but without further details a definite diagnosis can not be confirmed. Probably more common in the past were other diseases like infections (for example Pott’s disease caused by ), rickets, congenital and traumatic deformities 6 and spondylotic kyphosis. In addition, other diseases and some rare phenomena can irreversibly stiffen the spine, for example the Japanese “porter disease” (Lastenträgerkrankheit) in persons who have carried heavy loads on their back for a long time [24]. 7 More certainty about the diagnosis can be obtained by the examination of the remains of old skeletons. Not only human skeletons were searched for signs of AS, but also the remains of animals. 8

Paleopathology in animals Many fossils dated 30 to 50 million years ago were investigated. erosions or fusion, suggestive for spondyloarthritis, were seen in 13-50% of three 9 affected mammalian orders (like rhinocerotidae, to which the rhinos belong). Hypothesised was that persistence of this disease may perhaps represent evidence for some unknown host benefit [25]. In addition, more recent skeletons, extending 10 from the largest mammal that ever lived (the whale) to the one of the smaller marsupials, were also investigated for AS. Syndesmofytes and zygapophyseal joint fusion were found in cetacean (for example dolphins), erosions of zygapophyseal vertebral joints in a blue whale and sacroiliac joint fusion in a mouse-like marsupial [26]. Although a systemic inflammatory disease seems more likely than an infection

10 11 Chapter 1 Introduction

when polyarticular and symmetrical changes are seen, it remains doubtful what the However, the examined skeletons probably did not represent the average society, relation is between these changes in animals and human AS. Even more interesting because for example, the remains of seven early medieval bishops were included. are studies performed in human skeletons. The suggestion that a “monastic way of life” may have predisposed to DISH, was also supported in the study of the remains of presumably clergymen and high- Paleopathology in humans ranking citizens buried between 275 and 1795 in the abbey court in the city of 1 Interesting studies were done in remains of Egyptian pharaohs. Calcification of Maastricht. The mean age at death for adults was 37 years and in 40% DISH was spinal ligaments and ankylosis of the sacroiliac joints, suggestive of AS were diagnosed [33]. seen in mummies of Amenhotep II (reigned 1438-1412 BC), Ramses II (reigned about 1279-1213 BC) and, interestingly, his son Merenptah (reigned 1213-1203 BC) More indicative for AS were the skeletons of Cosimo de Medici (1389-1464), a banker 2 [27]. Ramses’s mummy also showed postinflammatory hip disease. His cervical in Florence, showing ossification and bridging of the vertebral ligaments and that of spine showed a post-mortem fracture of C5-6, probably produced intentionally to his son Piero, showing sacroiliac ankylosis [22, 34]. straighten his neck before mummification. The mummy of Thutmosis I (around 1500 These old skeletons were examined by inspection and , but more recently BC) showed an ankylosed spine, caused by another disease with hyperostosis: newer techniques became available. 3 diffuse idiopathic skeletal hyperostosis (DISH or Forestier’s disease, which is described below) [28]. Other authors stated that the findings in the skeletons of HLA-B27 in historical skeletons most of the mentioned pharaohs were also more suggestive for DISH than for AS Human Leucocyte Antigen (HLA) B27 is a strong genetic risk factor for AS. 4 [29, 30]. Investigated were remains from a church in La Neuville (Switzerland) of a male aged about 62 years living between the year 1300 and 1700. His spine showed extensive Bristol investigators reviewed 560 adult skeletons with at least part of the spine syndesmofytes, ossified interspinal ligaments and ankylosed facet joints, suggestive intact, gathered from recently excavated sites in England [31]. Most of them (424) of AS. B27 sequence-specific PCR (polymerase chain reaction) of DNA extracted 5 dated from the Middle Ages and in about half of the skeletons were from two parts of the femur showed confirmative results [35]. identified. Bilateral was seen in only one case and unilateral in another case, both with peripheral erosive joints changes more suggestive of another The skeleton from a church in Visby (Sweden) showed calcification of the vertebral 6 spondyloarthritis than AS. Thirteen specimens showed spinal signs of DISH. Hence, ligaments and fusion of intervertebral, costovertebral and sacroiliac joints, DISH seemed more prevalent despite the fact that people lived shorter and often compatible with AS. The man lived between the years 900 and 1300. HLA-B27 suffered from hunger in that period, whereas nowadays DISH is seen in middle-aged sequences were found in DNA, extracted from 20 mg of bone powder from several and older people and related to obesity and hyperglycaemia [32]. samples [36]. 7

8

9

10

Figure 2. Skeleton described by Bernard Connor (1695). Figure 3. Stretch-bed.

12 13 Chapter 1 Introduction

In both studies extensive precautions were taken to minimise the risk of Descriptions of AS in the nineteen’s century contamination. For example, several control samples were studied and genetic Carl Wenzel published in 1824 in Frankfurt-am-Main (Germany) findings in typing was done as well of all researchers involved in the first study. pathological anatomical specimens of spines with spondylotic and spondylitic ankylosis [37]. Patients with pain, stiffness and ankylosed spine were described for Early clinical descriptions of AS example by Lyons (1831), Hare (1849), Wilson (1856), Adams (1857) and Von Thaden 1 Realdo de Colombo, assistant of Andreas Vesalius and later professor of anatomy in (1863) [20]. Padua, published in 1559 anatomical descriptions of two skeletons typical of AS in Brodie described a patient in 1850 having gonarthritis and iritis as well, which might his book “De Re Anatomica” [23]. nowadays probably be classified as reactive arthritis. The two cases of Brodhurst (1858) had preceding gonorrhoea [20]. Several published clinical 2 Commonly accepted as the first most extensive description of AS are the reports pathological findings, suggestive of AS from the eighties (Fagge 1877, Sturge 1879) of Bernard Connor, an Irish medical student attending several medical schools in [20]. France and later personal physician of the King of Poland. He described in 1693 a However, in Garrod’s textbook from the year 1890 rheumatic fever was the most skeleton, which was found in a graveyard, in his letter to the British Royal physician important subject and particularly was seen as a separate disease [40]. The 3 Sir William de Waldegrave (figure 2) [37, 38]. “All these , which naturally are difference between , rheumatoid arthritis (RA) and AS was not yet separate and distinct from one another were here so straightly and intimately clear. RA was seen as a chronic sequel of rheumatic fever and its “development or joined, their ligaments perfectly bony and their articulations so effaced, … The root recrudescence is greatly favoured by exposure to damp and cold”. Rheumatic fever 4 of the ribs made by one equal smooth and plain superficies with the vertebrae and RA “can involve all joints, also the symphysis pubis, sacroiliac synchondroses and their apophyses… But when I had sawed two of the vertebrae asunder at the and the joints of the lumbar spine”. commissure I found that this uniting did not enter above two lines deep… their middles were separated as they usually are and touched each other only at their In the last decade of the century the most well-known early descriptions of AS were 5 edge…” [39]. published by Bechterew (in 1892-99), Strümpell (in 1884-97) and (in Somewhat amusing are his aetiological ideas: “As to the crooked and bending shape 1898). of the skeleton it is reasonable to suppose that it proceeded from the first formation 6 of the foetus in the womb, from the eggs not having sufficient room, or being Wladimir Michailowitsch Bechterew (1857-1927) was a Russian neurologist and accidentally pressed by some abscess in the womb or elsewhere, so that the carina psychiatrist. Some are convinced he was murdered, commissioned by Joseph of the backbone instead of running straight, was bent into a circle and kept the same Stalin several days after he had diagnosed Stalin as having paranoid psychosis [41]. figure when at full growth…” [20]. Bechterew described patients with “Steifigkeit der Wirbelsäule und Verkrümmung” 7 and several neurological signs like , neuralgia and muscle weakness. He assumed that the primary lesion was located in the meninges, descending from the neck to caudal. It is not clear why in several countries especially the name of 8 Bechterew was associated with the disease AS, because the cases he described where not the most classic ones of the disease as we know nowadays [42]. Adolf Strümpell (1853-1925) was a German neurologist. He described an illness in which, very gradually and without pain (!), from caudal to cranial the entire vertebral 9 column and hips became ankylosed completely [43]. Pierre Marie (1853-1940) was a French neurologist and pupil of Charcot. He described six male cases with spines “rigide comme un baton”, supplemented with 10 photographs and analysis of costovertebral specimens. The cases described by Strümpell and Marie were soon considered to be caused by the same disease (the Strümpell-Marie type of ) [22, 43].

Figure 4. Correction with pelottes (Barwell).

14 15 Chapter 1 Introduction

In the Netherlands in 1898 two cases from the Amsterdam clinic of were was an infective form of . Only five of the described patients were published in a Dutch magazine for and neurology by Jacobi and Wiardi female. Emphasizing the male predominance he noted that one of these women had Beckman [44]. In 1899 the Amsterdam family practitioner Martein Menko presented a calcified ovary and another was “somewhat masculine”, but the other three were for the society of physicians in Amsterdam a case of “ rhizomelica”, “true feminine”. Thirteen patients suffered from “iritis rheumatica”, which we now illustrated with photographs. He also quoted a case of André Leri in which the consider as anterior uveitis, strongly associated with AS. The characteristic posture 1 guard of the room was frightened by the sudden sight of a corps in of the patient was called “a sitting trunk on standing legs”. Interestingly, this typical sitting position. After the spine was dissected, the widow was surprised by the lying posture was noted especially in patients with physically light jobs. On the contrary, position of the corps, while she just ordered to make a coffin particularly constructed later more strenuous work was considered as a risk factor for worse prognosis [49]. for a sitting position [44]. 2 Jacques Forestier, a French specialist in internal diseases, described in 1950 the Was the medical world at the beginning of the new century ready to accept AS as a “senile ankylosing hyperostosis of the spine”, which was named after him as distinct disease? Forestier’s disease and was later called diffuse idiopathic skeletal hyperostosis (DISH). Cases were already reported since 1897 under names like “heredo-traumatic 3 AS in the first half of the twentieth century: the invention of X-rays. kyphosis of Bechterew” and “moniliform hyperostosis”. The distinction with AS was In addition to Bechterew’s disease several other names were used for illnesses that not fully accepted until then. However, Forestier showed the differences with AS were initially not always thought to be identical, like kyphosis heredo-traumatica, in age of onset, symptoms, clinical and radiological examinations. Most important 4 syndesmitis ossificans, spondylose rhizomélique, chronic ankylosing inflammation Forestier’s disease is a non-inflammatory disease of old age and radiologically the of the large joints and , Marie-Strümpell disease, rheumatoid “candle-flame” hyperostoses are entirely different from the syndesmofytes in AS spondylitis, spondylitis deformans and spondylarthritis ankylopoëtica. In 1905 the [28]. assumed differences between these diseases were also discussed in the Netherlands 5 by W Huet, neurologist/psychiatrist in Haarlem [45]. Jan van Breemen (1874-1961) was the pioneer of in Amsterdam and in the Netherlands. He was also one of the founders of the International In 1903 Eugen Fraenkel, a German pathologist, differentiated AS from osteoarthritis Organisation for the Investigation of Rheumatism (ICIR) in 1919, which evolved 6 by anatomic dissection. He also noted the involvement of the costovertebral joints in into the International League Against Rheumatism (ILAR) in 1925 and the European AS [43]. League Against Rheumatism (EULAR) in 1947 [50, 51]. In his Dutch textbook from Gradually X-rays became available for diagnosis and differentiation. The German the year 1942 Jan van Breemen emphasized the importance to diagnose AS in an Wilhelm Röntgen, who grew up in the Netherlands, developed X-rays in 1895. early phase [39]. He noted that patients could have symptoms for 10-20 years before 7 Already in 1897 spondylitis was studied radiographically [43]. In 1934 Walter Krebs, diagnosis. For early recognition he advised to detect the following early symptoms: a German radiologist, pointed out the significance of radiologic examination of the pain in spine and , X-ray anomaly of sacroiliacal joints, very limited chest sacroiliac joints and of signs of periostitis on the pelvis [43]. expansion, limited mobility of spine and elevated erythrocyte sedimentation rate 8 (ESR). Apparently, until then patients were often not diagnosed before they were Ralph Pemberton (USA) distinguished in his textbook (1929) two types: Bechterew quite ankylosed. and Marie [46]. “However, there seems to be no adequate reason to regard the spondylitides as representing anything more than the extension of the atrophic or Another European rheumatologist, Copeman (from Great Britain) agreed in his 9 hypertrophic type of arthritis to the vertebral column and associated structures.” textbook (1948) with Jan van Breemen that AS should be considered as a separate In the 1938 edition, Osler’s textbook (USA) dealt with rheumatoid spondylitis as one disease, in contrast to their colleagues in the USA [52]. form of chronic arthritis which included both osteoarthritis and RA, both included 10 under the term “arthritis deformans” [47]. From the second half of the twentieth century: HLA-B27 and classification criteria. More and more, AS was recognized as a disorder distinct from RA, because In 1936 Henri Tempelaar made a review of 54 cases of “spondylosis rhizomelica”, clinicians observed differences in age, sex, peripheral joint involvement, uveitis, examined and treated in the Consulting Bureau at Amsterdam (Medical Director: agglutination of sheep erythrocytes (Rose-test, the old test for ), Dr. Jan van Breemen) [48]. The writer shared the opinion of others that the disease response to chrysotherapy (gold) and response to radiotherapy [53].

16 17 Chapter 1 Introduction

However, even in 1956 in American literature AS was considered as an extension its scope to all aspects of SpA [68]. According to the new criteria it is possible to be of chronic RA, despite the fact that for years physicians had noticed another classified as having so called “non-radiographic axial SpA”, based on HLA-B27 and characteristic of AS: a hereditary factor [20]. Already in 1912 the appearance of clinical symptoms, even without MRI changes. This raises new questions, as for similar complaints in family members of AS patients was described by G Bolten, example about the prediction of the long term outcome of non-radiographic SpA. neurologist in Den Haag. [54]. Not all cases with non-radiographic SpA will develop to sacroiliitis in the end, as 1 In 1949 and 1951 Jan de Blécourt, from Groningen, reported that 6-19% out of 116 was recently shown [69]. Bechterew patients had one or more relatives with the same disease [55, 56]. In 1961 he published his investigation of 7405 family members of patients with RA or AS The recognition of AS, the concept of SpA, the international criteria and the and controls. The relative risk for having AS was 22.6 in families of AS patients and ASAS helped to come nearer to the ultimate goal: to improve the well-being and 2 for having RA 2.8 in families of RA patients [57]. outcome of patients with SpA [68]. This is pursuited by a tremendous amount of In 1973 the high association of the W27 Human Leucocyte Antigen (HLA) with AS investigations in SpA all over the world, in particular in epidemiology, etiology, was shown by Schlosstein [43]. Not much later, it was shown that HLA-B27 (as it was pathogenesis, genetics, pathology, radiology, clinical outcome, etcetera, and last but named) was also associated with anterior uveitis (iritis) and reactive arthritis [58, 59]. not least in therapy. 3 It also explained the higher prevalence of AS in northern countries, as HLA-B27 is In the next paragraph an overview of treatment of AS during the past hundred years more frequent there [2]. Although HLA-B27 is still the most important genetic factor is given. in AS, in later years it became clear that many other genetic factors are involved [4, 4 60]. Table 1. Modified New York criteria for ankylosing spondylitis (AS) (1984)

With the discovery of the strong association with HLA-B27, AS was definitely Clincal criteria: accepted as a separate disease. A disease needs a definition or criteria, particularly 5 Low and stiffness for more than 3 months which improves with exercise, but is when symptoms or signs are not pathognomonic. The first diagnostic criteria for not relieved by rest. AS were published after a meeting in Rome (1963) and the New York criteria were Limitation of motion of the lumbar spine in both the sagittal and frontal planes. formulated in 1966. In 1984 these criteria were revised to the now commonly used Limitation of chest expansion relative to normal values corrected for age and sex. 6 modified New York criteria (table 1) [61]. Radiologic criterion: In 1974 Moll and Wright introduced the term seronegative spondarthritis [62]. Sacroiliitis grade 2-4 bilaterally or grade 3-4 unilaterally. Several diseases that had been segregated from RA as separate entities in the preceding decades were now grouped together as spondarthritides, because of their Definite AS if the radiologic criterion is associated with at least one clinical criterion. 7 clinical, radiological, serological and genetic resemblances. It concerned especially AS, psoriatic arthritis, reactive arthritis and arthritis in inflammatory bowel disease [63]. Classification criteria for SpondyloArthritis (SpA) were published by Amor in 8 1990 and in 1991 by the European Study Group (ESSG) [64].

Before World War II limited chest mobility was considered as a major clinical symptom to diagnose AS in the “praespondylitic phase” (X-ray abnormalities 9 limited to the sacroiliacal joints) and called an important “Frühsymptom” [48, 65]. In later years, the SpA criteria covered the whole spectrum of SpA, including the true early phase of the disease without chronic X-ray changes. Recently, Magnetic 10 Resonance Imaging (MRI) made it possible to show active inflammation before X-ray structural damage become visible [66]. Therefore MRI was incorporated in the classification criteria for axial SpA by the Assessment of SpondyloArthritis international Society (ASAS) in 2009 (table 2) [67]. ASAS was initiated in 1995 to bring evidence-based unity in the existing assessments in AS and later broadened

18 19 Chapter 1 Introduction

Table 2. ASAS criteria for classification of axial spondyloarthritis (SpA) 3 HISTORY OF THERAPY IN ANKYLOSING SPONDYLITIS. In patients with back pain ≥ 3 months and age at onset < 45 years General and various treatment Sacroiliitis on imaging* OR HLA-B27 In the time that the discrimination between ankylosing spondylitis (AS) and plus plus rheumatoid arthritis (RA) was not yet accepted or not always possible, the same 1 were used for both diseases. Gradually differences in response, but also ≥ 1 SpA feature** ≥ 2 other SpA features similarities between these diseases became obvious. * Sacroiliitis on imaging: Active (acute) inflammation on MRI highly suggestive of sacroiliitis associated with SpA. OR In the decades after the first descriptions of AS the following therapies were most 2 Definite radiographic sacroiliitis according to modified New York criteria. decribed: salicylates, potassium iodide, antipyrin, arsenic, strychnin, application of iodine ** SpA features: - inflammatory back pain tincture above the vertebra with a pencil, corsets and several forms of physiotherapy 3 - arthritis [44, 70] Rest, diet, cod-liver oil and tonic measures were meant to raise the general - (heel) resistance and improve the calcium metabolism [52]. Other forms of therapies - uveitis included cupping (local suction on the skin by a device), vaccination (polyvalent - immune stimulation by for example Warren Crowe’s or Pondorff’s vaccin) and - psoriasis 4 - Crohn’s disease, ulcerative colitis treatment with testosteron (neo-hombreol syntheticum) [39, 71]. - good response to NSAIDs Following observations of increased calcium levels in AS, hemi-parathyroidectomy - family history for SpA seemed to give short term, but temporarily improvement in two-third of 49 operated - HLA-B27 patients [47, 48]. 5 - elevated C-reactive protein (CRP) In later years patients often used alternative or complementary , but in AS hardly any studies were done. In a study with simplified methods in 141 AS patients no evidence was found for Ayurvedic treatment (old traditional Indian healing 6 method) in AS [72]. Since decades a low starch diet is advocated by Alan Ebringer, based on his hypothesis that Klebsiella bacteria in the gut are involved in triggering AS. However, the efficacy of this treatment was never proven [73]. 7

8

9

10

Figure 5. Violent mobilisation (Beely).

20 21 Chapter 1 Introduction

Physiotherapy muscles in order to improve the kyphotic posture, was probably never applied [48]. Massage was used frequently, but in the past discouraged in case of active Braces were used for relaxation, but long-term negative effect on muscle strength inflammation [39]. was emphasized [48]. A corset was also used for correction by using a hinged Abnormal posture and limited mobility were supposed to be due to the “sleeping” plaster. The patient had to tighten the swivel every day for a bit in order to correct condition of the muscles, because patients avoided to move. Therefore, exercises his posture. Improvement of 10-14 degrees of kyphosis and 2-3 cm increase in body 1 were considered as very important. Chest exercises, for instance, could be length was reported in 1947 (figure 11) [79]. supported by a tight pelotte on the abdomen (Krebs’ method of coercion) and a sling was supposed to be used for passive of the spine (figures 3-6) [48, 74]. From 1945 on corrective operative treatment was reported by means of vertebral Jan van Breemen, however, concluded that gymnastics were not successful in osteotomy (the Norwegian-born Marius Smith-Petersen) [79]. In case of disabling 2 correcting the posture of the spine [39]. In contrast to exercises, rest was advocated kyphosis lumbar and sometimes cervical wedge osteotomy can still be an by others in 1948: “The patient should spend the greater part of the day in bed or advantageous treatment in highly specialised orthopaedic centres [80]. lying on his back” [52]. In addition, the technique of joint replacement by a prosthesis made it possible to For many years several types of physiotherapy were used, such as hydrotherapy, treat secondary osteoarthritis of the hip, which often partly corrects the kyphotic 3 spa, (Priessnitz’s) cold packs, methylsalicylate containing liniments, application of posture of the AS patient (figure 1). Total hip replacement is performed since 1966 warmth, light and electricity (figures 7-10). Temporary symptomatic relief of AS was (McKee, Watson-Farrar and Charnley) [81-83]. achieved by exposition to a quartz lamp, an ultraviolet light source, twice a week [39, 4 75]. Radiotherapy Since 1987 several controlled trials of individualized and group in Radioactive elements were used, like radon (product of radium) in hydrotherapy AS were reported and showed favourable effects on mobility, fitness and function and thorium (mesothorium X) as weekly injections. In recent years, just before the [76-78]. breakthrough of the Tumour Necrosis Factor-alpha (TNF) blockers, radium was again 5 applied as intravenous injections in Germany [84, 85]. Orthopaedic therapy Strümpell applied forced correction of spine and joints in deep narcosis has been used for several rheumatic diseases and seemed 6 (redressement forcé), associated with fracturing of ankylosed ligaments. It resulted especially efficacious in AS [86]. In fact, the existence of an effective treatment was in temporary excruciating pains and high fevers [44, 79]. the reason early diagnosis of AS was advocated. From about 1935 to 1955 radiation Dissection of the spinous processes and curettage of the sacroiliacal joints was was the most important therapy for AS when the effect of physiotherapy and also attempted, but not successful. The suggestion to disrupt the rectal abdominal 7

8

9

10

Figure 6. Sling suspension to stretch flexors of the spine. Figure 7. Bath with watermassage (Lahmann).

22 23 Chapter 1 Introduction

antiflogistics was insufficient. It was supposed that X-rays had an anti-inflammatory Antiflogistics effect and slowed down the ossification process [39, 71, 87]. From old times salicine, extracted from the bark of willow (Salix), has been used to The results of X-ray therapy in a small set of patients from the “Amsterdamsche reduce symptoms of inflammation [92]. Salicylate was isolated from salicin in 1838 Consultatiebureau voor Rheumatische ziekten van de Nederlandsche Vereeniging and came into use in rheumatology in the late 1870s [93, 94]. Several salicylates tot Rheumatiekbestrijding”, directed by Jan van Breemen, were published in 1940 were used in AS, for example phenyl salicylate (Salol) 3 gram a day orally and 1 [88]. Twentyseven patients aged 19-50 years received a total dose of 200-2800 rad acetyl-salicylacid (Aspirin) 3-6 gram a day, for their analgetic and anti-inflammatory (= 2-28 gray), divided over 4 fields of the total spine in series of about ten days. The effects [39, 44]. duration of follow-up was maximal 2 years. Outcome measures were subjective From 1929 dimethyl-amido-antipyrine (pyramidon) was available and was used in all improvement, mobility, erythrocyte sedimentation rate (ESR) and, interestingly, inflammatory rheumatic disorders in doses of 1500-4000 mg per day. Related drugs 2 ability to resume work. Fifteen patients showed large improvement (especially in were metamizol (known as novalgin, novaminsulfon and dipyrone) and from 1952 those with a disease duration of less than one year), 18 improvement of ESR, only (butazolidin; 200-1200 mg a day, mostly 400 mg) and irgapyrine 4 improvement of mobility and 8 patients were lost for follow-up. Fourteen patients (combination of pyramidon and phenylbutazone) [39, 95]. were able to resume their work, but this figure might have been influenced by socio- Most important adverse events of these antiflogistics, as in newer nonsteroidal 3 economic factors. anti-inflammatory drugs (NSAIDs), were gastro-intestinal ulceration, hypertension, A placebo-controlled (!) trial to study the effect of röntgen therapy in “rheumatoid fluid retention, nephrotoxicity and hemorrhagic diathesis. Life threatening was the spondylitis” (reported by Smith in 1947) showed a favourable effect, but details of occurrence of agranulocytosis in the group of pyrazolones, which was reported 4 the results are lacking [89]. from 1933 [95, 96]. Fifty years later, in 1983 this led to a medical debate in the In an open study in 1953 with 788 patients who had several rheumatic disorders, Netherlands about the risks and benefits of phenylbutazone [97-99]. The prescription adverse events of X-ray therapy were not mentioned [86]. However, others of phenylbutazone became limited to rheumatologists for treatment of AS, because sometimes described adverse events such as a hangover or loss of hair in the neck. phenylbutazone seemed to have a special role in this disease and the risks could be 5 More importantly, from the late forties lymphoproliferative disorders (aplasia and limited in this way [100, 101]. leukaemia) were reported in irradiated patients [90]. It was even stated that patients However, studies directly comparing different NSAIDs were scarce or included small had received a mean bone marrow radiation dose of about double that of atomic numbers of patients. Most of these studies showed comparable efficacy [94, 102, 6 bomb survivors [10]. Because these risks and the temporarily efficacy, radiation 103]. therapy was used only incidentally since 1965, but was still used in some countries Phenylbutazone is still being used in AS, but gradually newer NSAIDs were in the seventies [85, 87, 91]. used more frequently and from 1999 the COX-2 selective inhibitors (COXIBs) were introduced, because of their lower risk for gastric ulceration [104]. One of 7 the COXIBs, rofecoxib (Vioxx), was withdrawn in 2004 because of an increased 8

9

10

Figure 8. Alternating hot and cold shower. Figure 9. Steam-bath in bed.

24 25 Chapter 1 Introduction

cardiovascular risk [105]. Thereafter, it was discussed that many NSAIDs might In the late 1930s (SSZ) was designed by Nanna Svartz, in Stockholm, increase this risk, probably with the exception of [106]. for treatment of RA, combining the antibacterial properties of sulfonamide with the anti-inflammatory properties of salicylate [115]. However, the results of a small In 1976 it was shown in a retrospective study that, apart from the beneficial effects controlled study in 1948 seemed disappointing and SSZ became primarily popular on pain and stiffness, continuous use of phenylbutazone also seemed to delay in treating inflammatory bowel diseases. In the late 1970s a revival for the interest 1 radiological ossification of the lumbar spine [107]. Interestingly, recent controlled in SSZ for RA was seen and in the 1980s also in AS small placebo-controlled studies studies confirmed radiographic retardation by continuous NSAID use in contrast to were done, showing moderate effects [115, 116]. Later, larger studies were done and on-demand NSAID use, especially in patients with elevated acute phase reactants in 1995 SSZ proved to have favourable effect in AS with peripheral arthritis [117, 118]. [108-110]. In this subgroup of AS SSZ 2000-3000 mg a day is still indicated [119]. 2 Therefore, it seems that in addition to symptom modification, NSAIDs might also have disease modifying properties. The question is whether other disease modifying Levamisole was studied in a small double-blind trial in a mixed group of antirheumatic drugs (DMARDs) are effective in AS. spondyloarthritides with some positive effects, but also often adverse events [120]. Results of studies with , azathioprine, thalidomide and pamidronate in 3 DMARDs AS were conflicting and in most instances these studies were small and open [121]. In the past, the difference between AS and RA was not always accepted and not Possibly methotrexate can have a role in peripheral arthritis in AS, but there is not always possible in case of peripheral arthritis and therefore, even chrysotherapy enough evidence to recommend this [122]. 4 (intramuscular gold injections) was employed in AS. However, soon it was generally accepted that gold was not effective in AS, in contrast to RA [39, 65]. Biologicals The results of a very small double-blind placebo-controlled trial with D-penicillamine Around 1985 blocking of the pro-inflammatory cytokine Tumour Necrosis Factor- in AS were also negative [111]. alpha (TNF) was studied in bacterial sepsis in animals, but in patients this therapy 5 was not successful for this indication [123]. Later, TNF-blocking proved to be effective Soon after the discovery of cortisone (by Nobel prize winner in 1950 Philip Hench, a in RA and the first experiences in AS were very encouraging [124]. rheumatologist), cortisone (compound E) and Adrenocorticotropic hormone (ACTH), From 2002 onwards TNF-blocking therapy proved to be efficacious in large 6 were tried in AS, but the effect was temporary because was given shortly multicentre double-blind placebo-controlled trials in AS [125]. This applied for [90, 93, 112]. Currently, are mainly locally used in AS for uveitis (eye the intravenously administered infliximab, as well as the subcutaneously given drops), enthesitis (injection) and as intra-articular injection, especially in peripheral etanercept, adalimumab, and certolizumab [126-130]. TNF-blocking joints and sometimes in the sacroiliac joints [91, 113]. A recent placebo-controlled in general has favourable effects on axial symptoms and inflammatory Magnetic 7 trial showed short-term response of a short treatment with a high oral dose prednisolone [114]. 8

9

10

Figure 10. Electric light-bath (Bilz). Figure 11. Corrective corset.

26 27 Chapter 1 Introduction

Resonance Imaging (MRI) lesions, as well as on fatigue, quality of life, peripheral 4 OBJECTIVES AND OUTLINE OF THIS THESIS. arthritis, enthesitis and on extra-articular manifestations like uveitis, psoriasis and colitis [131-134]. Objectives of this thesis The most important adverse events of TNF-blockers are bacterial infections. After As demonstrated in the last paragraph, for a long time treatment of ankylosing the first TNF-blockers were approved for clinical practice, it became clear that life- spondylitis (AS) consisted mainly of physiotherapy and non steroidal anti 1 threatening reactivation of latent tuberculosis did occur [135]. Furthermore, in RA inflammatory drugs (NSAIDs). In contrast to rheumatoid arthritis, most disease the risk for non-melanoma skin cancer seems to be increased [136]. modifying antirheumatic drugs (DMARDs) were not effective in AS. The first part of this thesis concerns studies done just before TNF blockers became available and About 60-80% of the AS patients respond to treatment with a TNF-blocker, but several drugs with potential DMARD activity in AS were investigated. 2 only 30% show a very good response. Therefore, other treatment options remain The availability of the tumour necrosis factor (TNF) blocking agents changed the desirable and other biologicals have been and are being studied [137]. Until now therapeutic possibilities in AS dramatically and the second part of this thesis covers inhibition of the interleukin-12/23 and interleukine-17 pathways show promising studies concerning different aspects of treatment with a TNF-blocking agent. results in AS (ustekinumab and ) [138, 139]. 3 Outline of this thesis After the acceptance of the concept of axial SpA and non-radiographic SpA, it was Chapter 1 includes the clinical description of AS and the historical perspectives of obviously questioned if TNF-blocking would be useful in these patients. The first the evolution of this disease entity and treatment options. 4 studies showed favourable results [130, 140-143]. On the basis of these studies TNF-blocking therapy is now approved in clinical practice for non-radiographic axial AS is associated with inflammatory bowel disease for which mesalazine is a SpA in some countries, provided that active inflammation is demonstrable by the registered therapy. Mesalazine has been used in AS with various results. The efficacy presence of bone oedema on MRI or increased C-reactive protein levels. of a mesalazine with different pharmacological characteristics was studied in AS and 5 the results of this open study are described in chapter 2. Currently, physiotherapy and NSAID therapy maintain the cornerstones of therapy in AS and axial SpA [119]. However, the disease can remain active even when this In AS, the cardiovascular risk is increased and statins (cholesterol lowering agents) 6 treatment is given for an adequate period of time. Fortunately, for those patients have shown anti-inflammatory effects in rheumatoid arthritis. The effect of the statin since around 10 years a successful treatment is available in the form of TNF-blocking rosuvastatin on AS disease activity and lipid profile was studied in an open study. agents. The results of this study are shown in chapter 3. 7 Some studies described in this thesis started just before the availability of TNF- Leflunomide has proven to be effective in rheumatoid arthritis as well as in psoriatic blocking therapy and some started thereafter. In the latter ones, several aspects of arthritis. In chapter 4 the results of a double blind, randomised, placebo controlled the TNF-blocking therapy were studied. In the next paragraph the aims and outline of study of the efficacy of leflunomide in AS are shown. 8 this thesis are described. TNF-blocking agents are effective in the majority of AS patients. However, in some patients they lack efficacy from the start or after a period in which the disease responded well. The discovery of against TNF-blocking agents in 9 rheumatoid arthritis made clear that these antibodies influence the efficacy. Chapter 5 shows the results of a study investigating the role of antibodies against infliximab in AS. 10

Increased liver enzymes were observed in several AS patients treated with a TNF- blocking agent. In a retrospective study, the incidence of increased liver enzymes in AS patients treated with etanercept was investigated and possible risk factors were studied. The results are reported in chapter 6.

28 29 Chapter 1 Introduction

The effectiveness of a therapeutic agent in AS is determined with international REFERENCES CHAPTER 1 accepted outcome measures. However, most of these assessments are based on patient questionnaires. It is well known that patient-based and objective outcome [1] Braun J, Sieper J. Ankylosing spondylitis. Lancet 2007;369:1379-90. measures can differ. As part of a project to compare functional outcome on basis of [2] Khan MA. Update on . Ann Intern Med 2002;136:896-907. questionnaires and performance-based functional tests, the change in performance [3] Bakland G, Nossent HG, Gran JT. Incidence and prevalence of ankylosing spondylitis in 1 was studied in AS patients who were treated with a TNF-blocking agent. In chapter 7 Northern Norway. Arthritis Rheum 2005;53:850-5. the results are shown. [4] Dougados M, Baeten D. Spondyloarthritis. Lancet 2011;377:2127-37. [5] Rosenbaum JT, Davey MP. Time for a gut check: evidence for the hypothesis that HLA-B27 Acute anterior uveitis is a frequent extra-articular manifestation of AS. The effect of predisposes to ankylosing spondylitis by altering the microbiome. Arthritis Rheum 2 TNF-blocking agents is favourable, but the efficacy might differ between the several 2011;63:3195-8. agents. Chapter 8 shows the results of a study in which the number of flares of [6] Chung HY, Machado P, van der Heijde D, D’Agostino M-A, Dougados M. Smokers in early uveitis is investigated before and after start of adalimumab. axial spondyloarthritis have earlier disease onset, more disease activity, inflammation and damage, and poorer function and health-related quality of life: results from the 3 DESIR cohort. Ann Rheum Dis 2012;71:809-16. In AS, like in other inflammatory rheumatic diseases, the incidence of osteoporosis [7] Spoorenberg A, van der Heijde D, de Klerk E, Dougados M, de Vlam K, Mielants H, et al. and fractures is increased. The effect of a TNF-blocking agent on bone mineral Relative value of erythrocyte sedimentation rate and C-reactive protein in assessment of density and vertebral fractures was studied and reported in chapter 9. disease activity in ankylosing spondylitis. J Rheumatol 1999;26:980-4. 4 [8] Boonen A. Socioeconomic consequences of ankylosing spondylitis. Thesis, Maastricht In chapter 10 an overview of the evolving treatment of AS is given. The results of 2002. the previous studies are summarized and various aspects of drug therapy in AS are [9] Zink A, Braun J, Listing J, Wollenhaupt J and the German Collaborative Arthritis Centers. discussed (with Dutch summary). Disability and handicap in rheumatoid arthritis and ankylosing spondylitis: results from 5 the German rheumatological database. J Rheumatol 2000;27:613-22. [10] Braun J, Pincus T. Mortality, course of disease and prognosis of patients with ankylosing spondylitis. Clin Exp Rheumatol 2002;20(Suppl.28):S16-22. 6 [11] Ward MM. Predictors of the progression of functional disability in patients with ankylosing spondylitis. J Rheumatol 2002;29:1420-5. [12] Robertson LP, Davis MJ. A longitudinal study of disease activity and functional status in a hospital cohort of patients with ankylosing spondylitis. Rheumatology 2004;43:1565-8. 7 [13] Peters MJL, van der Horst-Bruinsma IE, Dijkmans BAC, Nurmohamed MT. Cardiovascular risk profile of patients with spondylarthropathies, particularly ankylosing spondylitis and psoriatic arthritis. Semin Arthritis Rheum 2004;34:585-92. [14] Zeboulon M, Dougados M, Gossec L. Prevalence and characteristics of uveitis in the 8 spondyloarthropathies: a systemic literature review. Ann Rheum Dis 2008;67:955-9. [15] Vosse D, Landewé R, van der Heijde D, van der Linden S, van Staa T-P, Geusens P. Ankylosing spondylitis and the risk of fracture: results from a large primary care-based nested case-control study. Ann Rheum Dis 2009;68:1839-42. 9 [16] Van der Weijden MA, Claushuis TA, Nazari T, Lems WF, Dijkmans BA, van der Horst- Bruinsma IE. High prevalence of low bone mineral density in patients within 10 years of onset of ankylosing spondylitis: a systematic review. Clin Rheumatol 2012;31:1529-35. [17] Peters MJL, Visman I, Nielen MMJ, van Dillen N, Verheij RA, van der Horst-Bruinsma 10 IE, et al. Ankylosing spondylitis: a risk factor for myocardial infarction? Ann Rheum Dis 2010;69:579-81. [18] Szabo SM, Levy AR, Rao SR, Kirbach SE, Lacaille D, Cifaldi M, et al. Increased risk of cardiovascular and cerebrovascular diseases in individuals with ankylosing spondylitis. A population-based study. Arthritis Rheum 2011;63:3294-3304.

30 31 Chapter 1 Introduction

[19] Vorstermanbijbel. Sinte Lucas Euangelie, capittel 13. Antwerpen, 1531. [40] Garrod AE. A treatise on Rheumatism and Rheumatoid Arthritis. Charles Griffin and [20] O’Connell D. Ankylosing spondylitis: the literature up to the close of the nineteenth company, London 1890. century. Ann Rheum Dis 1956;15:119-23. [41] Keitel W. Ein Opfer Stalins? Wladimir Michailowitsch Bechterew (1857-1927). Z Rheumatol [21] Grmek MD. Deformities of the spine and rheumatic diseases in Greco-Roman art. In: 2002;61:201-6. Appelboom, ed. Art, history and antiquity of rheumatic diseases. Elsevier, Brussel 1987. [42] Leden I. Did Bechterew describe the disease which is named after him? A question raised [22] Dequeker J, De Vlam K. The history of ankylosing spondylitis. In: Dijkmans BAC, Van due to the centennial of his primary report. Scand J Rheumatol 1994;23:42-5. 1 Royen BJ. Ankylosing Spondylitis, diagnosis and management. Taylor & Francis, New [43] Benedek TG. How did ankylosing spondylitis become a separate disease? Clin Exp York/ London 2006. Rheumatol 2009;27(Suppl.55):S3-S9. [23] Appelboom T. An illustrated history of rheumatic diseases. Springer, Houten 2010. [44] Menko MLHS. Spondylosis rhizomelica. Ned Tijdschr Geneeskd 1900;44:606-16. 2 [24] Dihlmann W. Die versteifte Wirbelsäule: bildgebende Differentialdiagnose unter [45] Huet WG. Chronisch abnormaal verloopende en von Bechterew’s Berücksichtigung des akquirierten Hyperostose-Syndroms und der Lastenträgerkrankheit. symptomen-complex (ingezonden). Ned Tijdschr Geneeskd 1905;49:1766-8. Z Rheumatol 2000;59:220-32. [46] Pemberton R. Arthritis and rheumatoid conditions. Lea & Febiger, Philadelphia 1929. [25] Rotschild BM, Sebes JI, Rotschild C. Antiquity of arthritis: Spondyloarthropathy identified [47] Bywaters EGL. Historical perspectives in the aetiology of ankylosing spondylitis. Br J 3 in the Paleocene of North America. Clin Exp Rheumatol 1998;16:573-5. Rheum 1983;22(supll.2):1-4. [26] Rotschild BM, Naples V. Whale of a tale. Ann Rheum Dis 2005;64:1385-6. [48] Tempelaar HCG. Spondylosis rhizomelica (spondylarthritis ankylopoetica). Ned Tijdschr [27] Feldtkeller E, Lemmel E-M. Ankylosing spondylitis in the pharaohs of ancient Egypt. Geneesk 1936;80:1898-1905. Rheumatol Int 2003;23:1-5. [49] Ward MM, Weisman MH, Davis JC Jr, Reveille JD. Risk factors for functional limitations 4 [28] Forestier J, Rotes-Querol J. Senile ankylosing hyperostosis of the spine. Ann Rheum Dis in patients with long-standing ankylosing spondylitis. Arthritis Care Res 2005;53:710-7. 1950;9:321-30. [50] Barnes CG, editor. The history of EULAR 1947-2007. EULAR, Kilchberg 2007. [29] Chhem RK, Schmit P, Fauré C. Did Ramesses II really have ankylosing spondylitis? A [51] Keitel W. Jan van Breemen (1874-1961). Z Rheumatol 2013;72:491-9. reappraisal. Can Assoc Radiol J 2004;55:211-7. 5 [52] Copeman WSC. Textbook of the rheumatic diseases. Livingstone, Edinburgh 1948. [30] Saleem SN, Hawass Z. Ankylosing spondylitis or diffuse idiopathic skeletal hyperostosis [53] Van Loghem-Langereis PE, van Loghem JJ. Drie serologische reacties bij rheumatische (DISH) in Royal Egyptian mummies of 18th -20th Dynasties? CT and archaeology studies. ziekten. Ned Tijdschr Geneeskd 1949;93:1983-8. Arthritis Rheumatol 2014;66:3311-6. [54] Bolten GC. Een geval van familiair voorkomende spondylosis rhizomelica (casuïstische 6 [31] Rogers J, Watt I, Dieppe P. Palaeopathology of spinal osteophytosis, vertebral ankylosis, mededelingen). Ned Tijdschr Geneeskd 1912;56:389-93. ankylosing spondylitis, and vertebral hyperostosis. Ann Rheum Dis 1985;44:113-20. [55] De Blécourt JJ. Familiair voorkomen van spondylarthritis ankylopoetica (ziekte van [32] Julkunen H, Heinonen OP, Pyörälä K. Hyperostosis of the spine in an adult population; its Bechterew). Ned Tijdschr Geneeskd 1949;93:3814-7. relation to hyperglycaemia and obesity. Ann Rheum Dis 1971;30:605-12. [56] De Blécourt JJ. Familiair voorkomen van spondylarthritis ankylopoetica (ziekte van 7 [33] Verlaan JJ, Oner FC, Maat GJ. Diffuse idiopathic skeletal hyperostosis in ancient Bechterew) II. Ned Tijdschr Geneeskd 1951;95:3763-6. clergymen. Eur Spine J 2007;16:1129-35. [57] De Blécourt JJ, Polman A, de Blécourt-Meindersma T. Erfelijke factoren bij reumatoide [34] Masi AT, Walsh EG. Ankylosing spondylitis: integrated clinical and physiological arthritis en spondylitis ankylopoetica. Ned Tijdschr Geneeskd 1961;105:1744-50. perspectives. Clin Exp Rheumatol 2003;21:1-8. 8 [58] Brewerton DA, Caffrey M, Nicholls A, Walters D, James DCO. Acute anterior uveitis and [35] Haak W, Gruber P, Rühli FJ, Böni T, Ulrich-Bochsler S, Frauendorf E, et al. Molecular HL-A 27. Lancet 1973;302:994-6. evidence of HLA-B27 in a historical case of ankylosing spondylitis. Arthritis Rheum 2005;52:3318-9. [59] Brewerton DA, Caffrey M, Nicholls A, Walters D, Oates JK, James DCO. Reiter’s disease and HL-A 27. Lancet 1973;302:996-8. [36] Leden I, Götherström A, Drenzel L, Svensson B. HLA-B27 sequences identified in a 9 mediaeval skeleton with ankylosing spondylitis. Ann Rheum Dis 2008;68:757-8. [60] Brown MA, Laval SH, Brophy S, Calin A. Recurrence risk modelling of the genetic susceptibility to ankylosing spondylitis. Ann Rheum Dis 2000;59:883-6. [37] Bywaters EGL. Historical introduction. In: Moll JMH, ed. Ankylosing spondylitis. Churchill Livingstone, Edinburgh, London, Melbourne and New York 1980. [61] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylits. Arthritis Rheum 1984;27:361-8. 10 [38] Keitel W. Der unternehmungslustige Ire Bernard Connor (1666-1698). Z Rheumatol 2012;71:330-9. [62] Wright V. Relationship between ankylosing spondylitis and other spondarthritides. In: Moll JMH. Ankylosing Spondylitis. Churchill Livingstone, Edinburgh, London, Melbourne [39] Van Breemen J. Chronische rheumatische ziekten. Van Holkema & Warendorf, Amsterdam and New York 1980. 1942. [63] Dekker-Saeys AJ. Spondylitis ankylopoetica-syndroom; een onderzoek naar het verband tussen spondylitis ankylopoetica en inflammatoire darmziekten. Thesis, Amsterdam 1976.

32 33 Chapter 1 Introduction

[64] Dougados M, van der Linden S, Juhlin R, Huitfeldt B, Amor B, Calin A, et al. The [84] Braun J, Lemmel EM, Manger B, Rau R, Sörensen, Sieper J. Therapie der European spondylarthropathy study group preliminary criteria for the classification of ankylosierenden spondylitis (AS) met radiumchlorid (224SpondylAT®). Z Rheumatol spondylarthropathy. Arthritis Rheum 1991;34:1218-27. 2001;60:74-83. [65] Dekkers HJN. Spondylosis rhizomelica (spondylarthritis ankylopoetica), samengaand met [85] Khan MA. Ankylosing Spondylitis, the history of medical therapies. Clin Exp Rheumatol arthritis der periphere gewrichten. Ned Tijdschr Geneeskd 1942;86:1010-6. 2002;20(Suppl. 28):S3-S5. [66] Oostveen JCM, Prevo R, den Boer J, van de Laar MAFJ. Early detection of sacroiliitis on [86] Desmarais MHL. Radiotherapy in arthritis. Ann Rheum Dis 1953;12:25-8. 1 magnetic imaging and subsequent development of sacroiliitis on plain radiography: a [87] Fellmann N. Die heutige Behandlung der Spondylitis ankylopoetica. Méd et Hyg prospective, longitudinal study. J Rheumatol 1999;26:1953-8. 1972;30:1953-5. [67] Rudwaleit M, van der Heijde D, Landewé R, Listing J, Akkoc N, Brandt J, et al. The [88] Van Ebbenhorst Tengbergen J, Dekkers HJN. Resultaten van de behandeling van lijders development of Assessment of SpondyloArthritis international Society classification aan spondylosis rhizomelica (spondylarthritis ankylopoetica) met röntgenstralen. Ned 2 criteria for axial spondyloarthritis (part II): validation and final selection.Ann Rheum Dis Tijdschr Geneeskd 1940;84:5010-9. 2009;68:777-83. [89] Smith RT, Boland EW, Hench PS. The effect of röntgen therapy in rheumatoid spondylitis. [68] Sieper J, Rudwaleit M, Baraliakos X, Brandt J, Braun J, Burgos-Vargas R, et al. The Ann Rheum Dis 1947;6:114-6. Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to 3 [90] De Blécourt JJ. Over spondylarthritis ankylopoëtica (ziekte van Pierre Marie-Strümpell- assess spondyloarthritis. Ann Rheum Dis 2009;68(Suppl.II):ii1-ii44. Bechterew) (clinische lessen). Ned Tijdschr Geneeskd 1950;94:506-13. [69] Kiltz U, Baraliakos X, Karakostas P, Igelmann M, Kalthoff L, Klink C, et al. Do patients with [91] De Blécourt JJ. De behandeling van spondylitis ankylopoetica (capita selecta). Ned non-radiographic axial spondylarthritis differ from patients with ankylosing spondylitis? Tijdschr Geneeskd 1971;115:19-22. Arthritis Care Res 2012;64:1415-22. 4 [92] Appelboom T. Arthropathy in art and the history of – through the [70] Huet WG. Stijfheid van de wervelkolom met verkromming. Ned Tijdschr Geneeskd centuries to cyclooxygenase-2 inhibitors. Rheumatology 2002;41(suppl.1):28-34. 1902;46:296-307. [93] Benedek TG. A century of American rheumatology. Ann Intern Med 1987;106:304-12. [71] Touw JF. Spondylarthritis ankylopoëtica; clinische les. Ned Tijdschr Geneeskd 5 1938;82:714-20. [94] Sieper J, Braun J, Rudwaleit M, Boonen A, Zink A. Ankylosing spondylitis: an overview. Ann Rheum Dis 2002;61(Suppl III):iii8-iii18. [72] Falkenbach A, Oberguggenberger R. Ayurveda in ankylosing spondylitis and . Ann Rheum Dis 2003;62:276-7. [95] Meyer MP. De toepassing van butazolidine bij rheumatoïde arthritis en bij de ziekte van Bechterew (oorspronkelijke stukken). Ned Tijdschr Geneeskd 1955;99:1546-53. [73] Braun J, Tak-Yan-Yu D. Ankylosing Spondylitis. In: Stone JH, ed. A clinician’s pearls and 6 myths in rheumatology. Springer, New York 2009. [96] Terwen AJL. Agranulocytose na pyramidon (referaten). Ned Tijdschr Geneeskd 1934;78:2223. [74] Wynn Parry CB, Deary J. Physical measures in rehabilitation. In: Moll JMH. Ankylosing Spondylitis. Churchill Livingstone, Edinburgh, London, Melbourne and New York 1980. [97] Offerhaus L. De Russische roulette van de pyrazolonen. Ned Tijdschr Geneeskd 1983;127:638-42. [75] Hulleman F. Het instituut voor physische therapie te Amsterdam. Elsevier 1905;15:234-42. 7 [98] Boersma JW. De Russische roulette van de pyrazolonen (ingezonden). Ned Tijdschr [76] Hidding A. Group physical therapy in ankylosing spondylitis. Thesis, Maastricht 1993. Geneeskd 1983;127:1167. [77] Dagfinrud H, Hagen KB, Kvien TK.Physiotherapy interventions for ankylosing spondylitis. [99] Hartman AP, van Oijen PLM, Bernaert ACMM, van Soesbergen RM, van der Leeden H, Cochrane Database Syst Rev 2008;1:CD002822. Hattink RO, et al. De Russische roulette van de pyrazolonen (ingezonden). Ned Tijdschr 8 [78] Masiero S, Bonaldo L, Pigatto M, Lo Nigro A, Ramonda R, Punzi L. Rehabilitation Geneeskd 1983;127:1168. treatment in patients with ankylosing spondylitis stabilized with [100] Boersma JW. Preventie van agranulocytose tijdens het gebruik van pyrazolon-derivaten, inhibitor therapy. A randomized controlled trial. J Rheumatol 2011;38:1335-42. in het bijzonder van aminofenazon (Pyramidon). Ned Tijdschr Geneeskd 1973;117:376-83. [79] Kuipers RKW. De orthopaedische behandeling van spondylarthritis ankylopoetica. Ned [101] Van der Korst JK. Gewrichtsziekten. Bohn, Scheltema & Holkema, Utrecht 1980. 9 Tijdschr Geneeskd 1947;91:851-6. [102] Van Gerwen F, van der Korst JK, Gribnau FWJ. Double-blind trial of naproxen and [80] Van Royen BJ. Lumbar osteotomy in ankylosing spondylitis. Thesis, Amsterdam 1999. phenylbutazone in ankylosing spondylitis. Ann Rheum Dis 1978;37:85-8. [81] Van der Houwen H. Artroplastiek van de heup (referaat). Ned Tijdschr Geneeskd [103] Calin A, Britton M. Sulindac in ankylosing spondylitis. Double-blind evaluation of 1967;111:1452. 10 sulindac and indomethacin. JAMA 1979;242:1885-6. [82] Sweeney S, Gupta R, Taylor G, Calin A. Total hip arthroplasty in ankylosing spondylitis: [104] Song IH, Poddubnyy A, Rudwaleit M, Sieper J. Benefits and risks of ankylosing outcome in 340 patients. J Rheumatol 2001;28:1862-6. spondylitis treatment with nonsteroidal anti-inflammatory drugs. Arthritis Rheum [83] Nystad TW, Furnes O, Havelin LI, Skredderstuen AK, Lie SA, Fevang BT. Hip replacement 2008;58:929-38. in patients with ankylosing spondylitis. Ann Rheum Dis 2014;73:1194-7.

34 35 Chapter 1 Introduction

[105] Bijlsma JWJ. Terugtrekking van rofecoxib: een signaal om voorzichtig te zijn met coxibs [122] Chen J, Veras MM, Liu C, Lin J. Methotrexate for ankylosing spondylitis. Cochrane bij patiënten met verhoogd cardiovasculair risico. Ned Tijdschr Geneeskd 2004;148:2162- Database Syst Rev 2013;2:CD004524. 4. [123] Beutler B, Milsark IW, Cerami AC. Passive immunization against cachectin/tumor necrosis [106] Trelle S, Reichenbach S, Wandel S, Hildebrand P, Tschannen B, Villiger PM, et al. factor protects mice from lethal effect of endotoxin. Science 1985;229: 869–71. Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis. [124] Van den Bosch F, Kruithof E, Baeten D, de Keyser F, Mielants H, Veys EM. Effects of a BMJ 2011;342:c7086. loading dose regimen of three infusions of chimeric monoclonal to tumour 1 [107] Boersma JW. Retardation of ossification of the lumbar vertebral column in ankylosing necrosis factor alpha (infliximab) in spondyloarthropathy: an open pilot study. Ann spondylitis by means of phenylbutazone. Scand J Rheumatol 1976;5:60-4. Rheum Dis 2000;59:428-33. [108] Wanders A, van der Heijde D, Landewé R, Béhier J-M, Calin A, Olivieri I, et al. [125] Khan MA. Ankylosing spondylitis and related spondyloarthropathies: the dramatic Nonsteroidal antiinflammatory drugs reduce radiographic progression in patients with advances in the past decade. Rheumatology 2011;50:637-9. 2 ankylosing spondylitis: a randomized clinical trial. Arthritis Rheum 2005;52:1756-65. [126] Van den Bosch F, Kruithof E, Baeten D, Herssens A, de Keyser F, Mielants H, et al. [109] Kroon F, Landewé R, Dougados M, van der Heijde D. Continuous NSAID use reverts Randomized double-blind comparison of chimeric to tumor the effects of inflammation on radiographic progression in patients with ankylosing necrosis factor alpha (infliximab) versus placebo in active spondylarthropathy.Arthritis spondylitis. Ann Rheum Dis 2012;71:1623-9. Rheum 2002;46:755-65. 3 [110] Poddubnyy D, Rudwaleit M, Haibel H, Listing J, Märker-Hermann E, Zeidler H, et al. Effect [127] Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosing spondylitis by inhibition of of non-steroidal anti-inflammatory drugs on radiographic spinal progression in patients tumour necrosis factor alpha. N Engl J Med 2002;346:1349-56. with axial spondyloarthritis: results from the German Spondyloarthritis Inception Cohort. [128] Van der Heijde D, Kivitz A, Schiff MH, Sieper J, Dijkmans BAC, Braun J, et al. Efficacy and 4 Ann Rheum Dis 2012;71:1616-22. safety of adalimumab in patients with ankylosing spondylitis: results of a multicenter. [111] Steven MM, Morrison M, Sturrock RD. Penicillamine in ankylosing spondylitis: a double Randomised, double-blind, placebo-controlled trial. Arthritis Rheum 2006;54:2136-46. blind placebo controlled trial. J Rheumatol 1985;12:735-7. [129] Inman RD, Davis JC Jr, van der Heijde D, Diekman L, Sieper J, Kim SI, et al. Efficacy and [112] Rübsaam CJ Jr. Behandeling van spondylitis ankylopoetica met cortison (referaat). Ned safety of golimumab in patients with ankylosing spondylitis: results of a randomized, 5 Tijdschr Geneeskd 1958;102:497. double-blind, placebo-controlled, phase III trial. Arthritis Rheum 2008;58:3402-12. [113] Maugars Y, Mathis C, Berthelot J-M, Charlier C, Prost A. Assessment of the efficacy of [130] Landewé R, Braun J, Deodhar A, Dougados M, Maksymowych WP, Mease PJ, et al. sacroiliac injections in spondylarthropathies: a double-blind study. Br J Efficacy of on of axial spondyloarthritis Rheumatol 1996;35:767-70. including ankylosing spondylitis: 24-week results of a double-blind randomised placebo- 6 [114] Haibel H, Fendler C, Listing J, Callhoff J, Braun J, Sieper J. Efficacy of oral prednisolone controlled phase 3 study. Ann Rheum Dis 2014;73:39-47. in active ankylosing spondylitis: results of a double-blind, randomised, placebo- [131] Targan SR, Hanauer SB, van Deventer S, Mayer L, Present DH, Braakman T, et al. A short- controlled short-term trial. Ann Rheum Dis 2014;73:243-6. term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for [115] Pinals RS. Sulfasalazine in the rheumatic disease. Semin Arthritis Rheum 1988;17:246-59. Crohn’s disease. N Engl J Med 1997;337:1029-35. 7 [116] Rains CP, Noble S, Faulds D. Sulfasalazine; a review of its pharmacological properties and [132] Elewaut D, Matucci-Cerinic M. Treatment of ankylosing spondylitis and extra-articular therapeutic efficacy in the treatment of rheumatoid arthritis. Drugs 1995;50:137-56. manifestations in everyday rheumatology practice. Rheumatology 2009;48:1029-35. [117] Dougados M, van der Linden S, Leirisalo-Repo M, Huitfeldt B, Juhlin R, Veys E, et al. [133] Baraliakos X, Braun J. Spondyloarthritides. Best Pract Res Clin Rheumatol 2011;25:825-42. Sulfasalazine in the treatment of spondylarthropathy; a randomized, multicenter, double- [134] Song IH, Hermann KG, Haibel H, Althoff CE, Listing J, Burmester GR, et al. Effects of 8 blind, placebo-controlled study. Arthritis Rheum 1995;38:618-27. etanercept versus sulfasalazine in early axial spondyloarthritis on active inflammatory [118] Chen J, Liu C. Is sulfasalazine effective in ankylosing spondylitis? A systemic review of lesions as detected by whole-body MRI (ESTHER): a 48-week randomised controlled trial. randomized controlled trials. J Rheumatol 2006;33:722-31. Ann Rheum Dis 2011;70:590-6. 9 [119] Braun J, van den Berg R, Baraliakos X, Boehm H, Burgos-Vargas R, Collantes-Estevez [135] Furst DE, Keystone EC, Braun J, Breedveld FC, Burmester GR, de Benedetti F, et al. E, et al. 2010 update of the ASAS/EULAR recommendations for the management of Updated consensus statement on biological agents for the treatment of rheumatic ankylosing spondylitis. Ann Rheum Dis 2011;70:896-904. diseases, 2011. Ann Rheum Dis 2012;71(Supp II):i2-i45. [120] Creemers MCW, van Riel PLCM, Franssen MJAM, van de Putte LBA, Gribnau FWJ. [136] Chakravarty EF, Michaud K, Wolfe F. Skin cancer, rheumatoid arthritis, and tumor necrosis 10 Second-line treatment in seronegative spondylarthrpathies. Sem Arthritis Rheum factor inhibitors. J Rheumatol 2005;32:2130-5. 1994;24:71-81. [137] Wendling D, Prati C. Biologic agents for treating ankylosing spondylitis: beyond TNF [121] Zochling J, van der Heijde D, Dougados M, Braun J. Current evidence for the alpha antagonists. Joint Bone Spine 2011;78:542-4. management of ankylosing spondylitis: a systemic literature review for the ASAS/EULAR [138] Baeten D, Baraliakos X, Braun J, Sieper J, Emery P, van der Heijde D, et al. Anti- management recommendations in ankylosing spondylitis. Ann Rheum Dis 2006;65:423- interleukine-17A monoclonal antibody secukinumab in treatment of ankylosing spondylitis: 32. a randomised, double-blind, placebo-controlled trial. Lancet 2013;382:1705-13.

36 37 Chapter 1 Introduction

[139] Poddubnyy D, Hermann K-GA, Callhoff J, Listing J, Sieper J. Ustekinumab for the treatment of patients with active ankylosing spondylitis: results of a 28-week, prospective, open-label, proof-of-concept study (TOPAS). Ann Rheum Dis 2014;73:817-23. [140] Haibel H, Rudwaleit M, Listing J, Heldmann F, Wong RL, Kupper H, et al. Efficacy of adalimumab in the treatment of axial spondyloarthritis without radiographically defined sacroiliitis: results of a twelve-week randomized, double-blind, placebo-controlled trial 1 followed by an open-label extension up to week fifty-two. Arthritis Rheum 2008;58:1981- 91. [141] Sieper J, van der Heijde D, Dougados M, Mease P, Maksymowych WP, Brown MA, et al. Efficacy and safety of adalimumab in patients with non-radiographic axial 2 spondyloarthritis: results of a randomised placebo-controled trial (ABILITY-1). Ann Rheum Dis 2013;72:815-22. [142] Song IH, Weiß A, Hermann K-GA, Haibel H, Althoff CE, Poddubnyy D, et al. Similar response rates in patients with ankylosing spondylitis and non-radiographic axial 3 spondyloarthritis after 1 year of treatment with etanercept: results from the ESTHER trial. Ann Rheum Dis 2013;72:823-5. [143] Sieper J, Lenaerts J, Wollenhaupt J, Rudwaleit M, Mazurov VI, Myasoutova L, et al. Efficacy and safety of infliximab plus naproxen versus naproxen alone in patients with 4 early, active axial spondyloarthritis: results from the double-blind, placebo-controlled INFAST study, part 1. Ann Rheum Dis 2014;73:101-7. 5

6

7

8

9

10

38 39 CHAPTER 2

Efficacy and safety of mesalazine (Salofalk®) in an open study of 20 patients with ankylosing spondylitis.

J C van Denderen1 I E van der Horst-Bruinsma2 P D Bezemer3 B A C Dijkmans1,2

Journal of Rheumatology 2003;30:1558-1560.

1 Jan van Breemen Instituut, Amsterdam. 2 Department of Rheumatology, VU University Medical Center, Amsterdam. 3 Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam. Chapter 2 Mesalazine in AS

ABSTRACT INTRODUCTION

Objective Treatment of patients with ankylosing spondylitis (AS) consists mainly of exercise Mesalazine (Salofalk®) was found to be effective and showed low toxicity in therapy and nonsteroidal antiinflammatory drugs (NSAID). A considerable patients with inflammatory bowel disease. The association of gut lesions and number of studies have assessed the efficacy and safety of sulfasalazine (SSZ) in 1 spondyloarthropathy (SpA) is well known and we studied the efficacy and safety of a AS [1-8]. SSZ (2–3 g/day) was proven to be more effective than placebo in active relatively high dose of mesalazine in patients with ankylosing spondylitis (AS). spondyloarthropathy (SpA), especially in decreasing the peripheral arthritis [6-8].

Methods SSZ is metabolized in the large intestine into sulfapyridine and mesalazine 2 In an open study, mesalazine (3–4 g/day) was prescribed for 24 weeks to 20 patients (5-aminosalicylic acid, 5-ASA). The latter is the active drug in the treatment of (aged 18–70 yrs) with active AS, defined as the presence of at least one clinical inflammatory bowel disease (IBD). Based on the hypothesis that the gut plays an criterion (morning stiffness > 30 min, peripheral synovitis, , or pain important role in the onset of AS and because mesalazine is less toxic than SSZ, 3 score > 2 on a visual analog scale of 10 cm) and one laboratory criterion [erythrocyte mesalazine seemed to be an attractive candidate in the treatment of AS. Mesalazine sedimentation rate (ESR) > 20 mm/h or C-reactive protein (CRP) > 20 mg/l]. Data was used previously in some AS patients with various results [9-11]. The results on toxicity and disease activity variables (ESR, CRP, BASDAI, BASFI, BASMI, global of a randomized controlled study of treatment with either SSZ, mesalazine, or assessment, and joint count) were obtained at baseline and after 4, 12, and 24 sulfapyridine suggested that sulfapyridine and not mesalazine is the active moiety in 4 weeks, and analyzed on an intention-to-treat basis. SpA. [11] However, in that study a very low dose of mesalazine (Asacol® 0.8 g/day) was used. For IBD, it is common practice to use doses up to 6 g/day of mesalazine. Results Our aim was to assess the efficacy and safety of a relatively high dose (up to Study patients had a mean age of 41 years, with mean disease duration of 7.9 years 4 g/day) of mesalazine (as Salofalk®) in patients with AS. 5 and a mean ESR at baseline of 29 mm/h. After a mean of 9.3 weeks (range 2–22), 8 of the 20 patients prematurely stopped the medication because of adverse effects, mainly gastrointestinal complaints. Twelve patients completed the 24 weeks of the MATERIALS AND METHODS 6 study using a mean dose of 3.2 g/day (range 1–4) mesalazine. Analysis of the data showed improvement in ESR, CRP, and physician’s global assessment, but only the Study design change in ESR (29 mm/h on baseline and 25 mm/h at week 24) reached statistical In an open pilot study, 20 patients with AS were treated with mesalazine over 24 significance (p = 0.03). No change was observed in the other disease activity weeks. Patients with active AS were included if they fulfilled the modified New 7 variables. York criteria for AS, were aged between 18 and 70 years, and showed active disease defined as the presence of at least one clinical criterion [morning stiffness Conclusion > 30 minutes, or peripheral synovitis, or enthesopathy, or pain score > 2 on a 8 No significant improvement in any disease activity variable of active AS was visual analog scale (VAS, 0–10 cm)] plus one laboratory criterion [erythrocyte observed during treatment with Salofalk® except for the ESR. Many side effects were sedimentation rate (ESR) > 20 mm/h or C-reactive protein (CRP) > 20 mg/l]. seen. Previous use of mesalazine, treatment with a disease modifying antirheumatic drug (DMARD), including SSZ, experimental therapy, or corticosteroids in the previous 4 9 weeks, known to salicylates or SSZ, pregnancy, severe renal and/or hepatic dysfunction, and history or symptoms of IBD were the exclusion criteria. After inclusion, mesalazine was prescribed as Salofalk® in 500 mg tablets with an 10 initial daily dose of 3 g (1 g tid). We confirmed with the patients’ that no other mesalazine formulation was given. In case of intolerance or side effects, the dosage of Salofalk® was decreased to the highest tolerated dose. After 4 weeks the daily dose was increased to 4 g in case of inefficacy, defined as less than 20% improvement in at least 2 of the following variables: VAS morning stiffness, VAS pain, or ESR.

42 43 Chapter 2 Mesalazine in AS

NSAID were continued during the study if they had been taken in a stable dose from Table 1. Number of reported adverse effects in 20 patients with AS taking mesalazine (adverse 4 weeks prior to study entry. The type, dosage at entry, and change in dosage or type effects were seen in 15 patients). of the NSAID during the study were recorded. Adverse Effect n At baseline and at 4, 12, and 24 weeks the following data were obtained by one of us GI disorders (JC van D): Bath Ankylosing Spondylitis Functional Index (BASFI), Bath Ankylosing Nausea 4 1 Spondylitis Disease Activity Index (BASDAI), global assessment according to the Abdominal pain 4 patient and to the doctor (VAS 0–10 cm), Bath Ankylosing Spondylitis Metrology Diarrhea 7 Index (BASMI, new scoring system) [12], tender joint score (TJS, 42 joints), swollen Increased hepatic enzymes 1 joint score (SJS, 40 joints), laboratory tests, and adverse events. Skin disorders 2 The following were used as variables for disease activity: ESR, CRP, BASDAI, BASFI, Pruritus 1 BASMI, global assessment according to the patient and to the doctor, TJS, and SJS. Worsening eczema 1 The Assessments in Ankylosing Spondylitis Working Group (ASAS) response criteria were not used because they were not developed when this study was performed CNS disorders 3 [13]. The mean values of the variables at baseline and after 24 weeks were compared Dizziness 3 using paired t tests. An intention-to-treat analysis of all 20 patients was performed, Headache 1 as well as a separate analysis of the patients who completed the whole study. Other 4 The medical ethical committee of the Slotervaart Hospital, Amsterdam, approved the Fever 1 study. Impotence 1 2 5 RESULTS GI: gastrointestinal, CNS: central nervous system.

The mean age of the 20 AS patients was 41 years (range 19–69, median 40), 18 were men and 19 were positive for the HLA-B27 antigen. The mean disease duration was All patients used NSAID during the study; 3 patients increased the dose, 4 used a 6 7.9 years (range 0.4–27). The patients had no history of extraarticular manifestations lower dose, and 3 switched to another NSAID. besides acute anterior uveitis (25%) and psoriasis (10%). The results of the disease outcome variables at baseline and during followup of Eight out of 20 patients stopped the medication permanently after a mean period the total group of 20 patients and of the 12 patients who completed the study are of 9.3 weeks (range 2–22) due to adverse effects, despite dose reduction. Several presented in Table 2. Improvement was observed in ESR, CRP, and physician’s global 7 of these patients were not even able to tolerate a dose as low as 0.5 g/day. assessment, but only the change in ESR reached statistical significance (p = 0.03). Most patients (75%) reported side effects, consisting mainly of gastrointestinal The other outcome variables did not change favorably. complaints, especially diarrhea. Five patients reported no adverse reaction due to 8 the medication (Table 1). Laboratory values showed no adverse effects except a A secondary analysis of the 12 completers only showed significant improvement of > 3–6-fold increase in hepatic enzymes in one patient, necessitating withdrawal of the CRP (p = 0.03) and physician’s global assessment (p = 0.02). treatment with mesalazine; the levels normalized after drug discontinuation. 9 Twelve (60%) of the patients completed the 24-week treatment with mesalazine, using a mean dose of 3.2 g/day (range 1–4). Because of adverse effects, 5 of these 12 patients discontinued the medication during a short period (with a mean duration of 10 3.2 weeks and in one case 10 weeks because of an intercurrent urological analysis), but completed the study. The 12 completers were younger (mean 34 yrs) and had shorter disease duration (mean 6.0 yrs) compared to the 8 dropouts (mean 52 and 10.7 yrs).

44 45 Chapter 2 Mesalazine in AS

Table 2. Disease activity variables in 20 patients with AS (data in parentheses for the 12 preparations of mesalazine. These various preparations are released at different completers only). Values are given as mean. parts of the bowel. The release of Pentasa® starts in the proximal small intestine, Baseline 4 Weeks 12 Weeks 24 Weeks p whereas Asacol® becomes available only when the pH rises to around 7, typically in ® ESR, mm/h 29 (30) 24 26 25 (27) 0.03 (0.26) the terminal ileum. Salofalk takes an intermediate place, being released at around CRP, mg/l 25 (30) 22 28 22 (23) 0.16 (0.03) pH 6. The mesalazine in SSZ is split from sulfapyridine in the colon by bacterial 1 BASDAI, 0–10 4.4 (4.4) 4.1 4.4 4.2 (3.7) 0.67 (0.31) enzymes. The associations of SpA with enterogenic infection and IBD are well known. Even in undifferentiated SpA and AS, ileocolonoscopic inflammatory lesions BASFI, 0–10 4.5 (4.1) 4.3 4.5 4.5 (3.8) 0.97 (0.49) in the small and large bowel were found in high frequencies [14]. However, because BASMI, 0–10 4.5 (3.9) 4.6 4.7 4.6 (4.2) 0.54 (0.09) the exact pathogenic role of the gut in AS is not known, the importance of the Patient global, 0–10 4.8 (4.9) 4.8 5.1 4.7 (4.3) 0.85 (0.35) 2 differences in delivery characteristics of the forms of mesalazine is not certain, but Doctor global, 0–10 5.4 (5.3) 4.8 4.8 4.6 (4.1) 0.09 (0.02) cannot be excluded. SJS, 0–40 0.2 (0.2) 0.2 0.3 0.3 (0.3) 0.33 (0.34) In summary, we saw no improvement in disease activity variables in patients with TJS, 0–42 1.2 (1.2) 1.2 1.5 1.3 (0.8) 0.80 (0.17) AS during treatment with Salofalk®, except for the ESR. There was a high rate of 3 Paired t test comparing week 24 to baseline. ESR: erythrocyte sedimentation rate, CRP: C-reac- premature discontinuance by patients because of intolerance. Although we cannot tive protein, BASDAI: Bath Ankylosing Spondylitis Disease Activity Index, BASFI: Bath Ankylos- exclude the possibility that a lower dose or a different formulation of mesalazine ing Spondylitis Functional Indesx, BASMI: Bath Ankylosing Spondylitis Metrology Index (new might be better tolerated and more effective, our results suggest that Salofalk® has scoring system), SJS: swollen joint score, TJS: tender joint score. 4 no role in the treatment of AS.

DISCUSSION 5

In our open pilot study, no significant improvement of active AS was observed during treatment with mesalazine for 24 weeks. The ESR was the only disease activity variable that changed significantly, although the clinical relevance of this 6 finding (mean difference –4 mm/h) was doubtful. A striking number of patients did not tolerate mesalazine, mainly due to gastrointestinal complaints, which improved after cessation of the drug. A separate analysis of the 12 patients who completed the 7 whole study did not show any favorable effect. Another preparation of mesalazine, Pentasa®, performed better in 2 open studies in patients with SpA, with fewer side effects and improvement of most clinical, physical, and laboratory variables [9,10]. The somewhat lower mean age of the 8 patients in these studies (37.6, 39.1, and 34.4 yrs) compared to our population (41 yrs) might be a possible explanation, because in our study the older patients were more likely to drop out. Also, the higher tolerance of Pentasa® can probably be 9 explained by the more gradual increment of the drug and the lower dose used in these studies. However, in some of our patients even a dose as low as 0.5 g/day in rechallenge was not tolerated. As far as we know there are no studies suggesting that Salofalk® is less tolerated 10 than other formulations of mesalazine. The impaired tolerance of mesalazine in our study, compared to the situation in IBD, is possibly related to the rheumatic disease itself or concomitant use of antiinflammatory drugs. It can be hypothesized that the apparent difference in efficacy between Pentasa® and Salofalk® could be due to the pharmacological difference between the several

46 47 Chapter 2 Mesalazine in AS

REFERENCES

[1] Dougados M, Boumier P, Amor B. Sulphasalazine in ankylosing spondylitis: a double blind controlled study in 60 patients. BMJ 1986;293:911-4. [2] Feltelius N, Hallgren R. Sulphasalazine in ankylosing spondylitis. Ann Rheum Dis 1986;45:396-9. 1 [3] Nissila M, Lehtinen K, Leirisalo-Repo M, Luukkainen R, Mutru O, Yli-Kerttula U. Sulfasalazine in the treatment of ankylosing spondylitis; a twenty-six-week, placebo- controlled clinical trial. Arthritis Rheum 1988;31:1111-6. 2 [4] Corkill MM, Jobanputra P, Gibson T, Macfarlane DG. A controlled trial of sulphasalazine treatment of chronic ankylosing spondylitis: failure to demonstrate a clinical effect. Br J Rheumatol 1990; 29:41-5. [5] Taylor HG, Beswick EJ, Dawes PT. Sulphasalazine in ankylosing spondylitis. A radiological, 3 clinical and laboratory assessment. Clin Rheumatol 1991;10:43-8. [6] Kirwan J, Edwards A, Huitfeldt B, Thompson P, Currey H. The course of established ankylosing spondylitis and the effects of sulphasalazine over 3 years. Br J Rheumatol 1993;32:729-33. 4 [7] Dougados M, van der Linden S, Leirisalo-Repo M, et al. Sulfasalazine in the treatment of spondylarthropathy; a randomised, multicenter, double-blind, placebo-controlled study. Arthritis Rheum 1995;38:618-27. [8] Clegg DO, Reda DJ, Weisman MH, et al. Comparison of sulfasalazine and placebo in the 5 treatment of ankylosing spondylitis; a Department of Veterans Affairs cooperative study. Arthritis Rheum 1996;39:2004-12. [9] Dekker-Saeys BJ, Dijkmans BAC, Tytgat GNJ. Treatment of spondyloarthropathy with 5-aminosalicylic acid (mesalazine): an open trial. J Rheumatol 2000;27:723-6. 6 [10] Thomson GTD, Thomson BRJ, Siwik Thomson K, Santos Ducharme J. Clinical efficacy of mesalazine in the treatment of the spondyloarthropathies. J Rheumatol 2000;27:714-8. [11] Taggart A, Gardiner P, McEvoy F, Hopkins R, Bird H. Which is the active moiety of sulfasalazine in ankylosing spondylitis? Arthritis Rheum 1996;39:1400-5. 7 [12] Jones SD, Porter J, Garret SL, Kennedy LG, Whitelock H, Calin A. A new scoring system for the Bath Ankylosing Spondylitis Metrology Index (BASMI) [letter]. J Rheumatol 1995;22:1609. 8 [13] Van der Heijde D, Calin A, Dougados M, Khan MA, van der Linden S, Bellamy N, on behalf of the ASAS Working Group. Selection of instruments in the core set for DC-ART, SMARD, physical therapy, and clinical record keeping in ankylosing spondylitis. Progress report of the Assessments in Ankylosing Spondylitis Working Group. J Rheumatol 1999;26:951-4. 9 [14] Mielants H, Veys EM, Cuvelier C, de Vos M, Botelberghe L. HLA-B27 related arthritis and bowel inflammation; II. Ileocolonoscopy and bowel histology in patients with HLA-B27 related arthritis. J Rheumatol 1985;12:294-8. 10

48 49 CHAPTER 3

Statin therapy might be beneficial for patients with ankylosing spondylitis.

J C van Denderen1 M J L Peters2 V P van Halm2 I E van der Horst-Bruinsma2 B A C Dijkmans2 M T Nurmohamed1

Annals of the Rheumatic Diseases 2006;65(5):695-696.

1 Department of Rheumatology, Jan van Breemen Instituut, Amsterdam. 2 Department of Rheumatology, VU University Medical Center, Amsterdam. Chapter 3 Statin therapy in AS

ABSTRACT Therapeutic options for patients with the chronic inflammatory disease ankylosing spondylitis (AS) are limited. Treatment was, until recently, mainly based on non- Objective. steroidal anti-inflammatory drugs and physical therapy. The efficacy of disease There is growing evidence that statins possess anti-inflammatory properties, as modifying antirheumatic drugs, such as sulfasalazine and methotrexate, is less indicated by lowering of C reactive protein levels (CRP), and the beneficial effect beneficial in AS than in other rheumatic diseases such as rheumatoid arthritis [1]. 1 of atorvastatin on disease activity in rheumatoid arthritis. Therefore, we conducted Recently, tumour necrosis factor-α blocking agents, have been proved to be very an open pilot study to investigate the effect of rosuvastatin on disease activity in effective in a high proportion of patients with AS. However, these agents are patients with ankylosing spondylitis (AS). expensive and their use is sometimes accompanied by severe adverse events, as opportunistic infections. Moreover, tumour necrosis factor-α blocking agents are 2 Methods not effective in about 30% of patients [2]. Hence, there is a continuing need for Fifteen unselected consecutive outpatients with active AS were treated with alternative therapeutic options. rosuvastatin (20 mg/day) for 12 weeks, followed by an observational phase of 12 3 weeks. Plasma concentrations of CRP, erythrocyte sedimentation rate (ESR), lipid There is growing evidence that statins possess anti-inflammatory properties, as levels and the following clinical measures: Bath Ankylosing Spondylitis Disease indicated by lowering of C reactive protein (CRP) levels, and recently, the clinically Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index, Bath beneficial effect of the statin atorvastatin on disease activity was demonstrated Ankylosing Spondylitis Metrology Index (BASMI), general wellbeing according to in rheumatoid arthritis [3, 4]. Therefore, we conducted an open pilot study to 4 the doctor/patient and pain score, were determined. investigate the effect of rosuvastatin on disease activity in patients with AS.

Results Fifteen unselected consecutive outpatients with AS, according to the modified New The mean age was 46 years (range 29–61) and the mean disease duration was 11.5 York criteria, were treated with rosuvastatin (20 mg/day) for 12 weeks, followed 5 years (range 2–22). All patients completed the trial, but the dose of rosuvastatin was by an observational phase of 12 weeks. Patients were eligible for inclusion if reduced in two patients owing to side effects. Clinical measures tended to improve they had active disease defined as at least four points on the Bath Ankylosing during the treatment period. Improvements in BASDAI, BASMI and pain score were Spondylitis Disease Activity Index (BASDAI, 0–10) and a pain score of at least four 6 sustained, with further improvement, during the observational phase. In addition, on the visual analogue scale (0–10). The local ethics committee approved the study treatment with rosuvastatin resulted in significant improvements of CRP and ESR protocol and all patients gave written informed consent. Main exclusion criteria after 12 weeks. Total and LDL-cholesterol were significantly reduced after 6 and 12 included cardiovascular events within the previous 3 months, current lipid lowering weeks and increased during the observational phase. treatment, and current use of biological agents. No patients were allowed to enter 7 the study who would otherwise have qualified for statins on the basis of calculated Conclusion risk. Treatment with rosuvastatin leads to an improvement of disease activity in patients Plasma concentrations of CRP, erythrocyte sedimentation rate (ESR), total 8 with active AS and is accompanied by significant reduction of acute phase reactants. cholesterol, low density lipoprotein (LDL)-cholesterol, high density lipoprotein Several clinical measures continued to improve during the follow up phase, (HDL)-cholesterol, triglycerides and the following clinical measures: BASDAI, Bath suggesting longlasting beneficial effects. Confirmatory randomised studies are Ankylosing Spondylitis Functional Index (BASFI, 0–10), Bath Ankylosing Spondylitis required. Metrology Index (BASMI, 0–10), general wellbeing according to the doctor/patient 9 (GWBD/P), and pain score were determined. Statistical analysis was performed with Wilcoxon’s signed rank test. Values of p<0.05 were considered significant. 10 The patients’ mean age was 46 years (range 29–61), nine were men and 14 were HLA-B27 positive. The mean disease duration was 11.5 years (range 2–22). All patients completed the trial. However, the dose of rosuvastatin was reduced in two patients to 10 mg/day after 6 weeks owing to side effects. Clinical measures, such as BASDAI, BASMI, pain score, and GWBD/P, tended to improve during the

52 53 Chapter 3 Statin therapy in AS

treatment period. Improvements in the first three variables were sustained, with Table 1. Clinical and biological variables (median) in 15 patients at baseline, at weeks further improvement, during the observational phase. In addition, treatment with 6, 12 (with rosuvastatin), and at weeks 18, 24 (without rosuvastatin). rosuvastatin resulted in significant improvements of CRP and ESR after 12 weeks. After stopping

Total and LDL-cholesterol were significantly reduced after 6 and 12 weeks and With rosuvastatin rosuvastatin increased during the observational phase (table 1). At 6 At 12 At 18 At 24 1 Variable Baseline weeks weeks weeks weeks The main conclusion is that treatment with rosuvastatin leads to an improvement ESR (mm/1st h) 22.0 20.0 15.0* 18.0 15.0 of disease activity in patients with active AS and is accompanied by significant CRP (log mg/l) 15.0 10.0 10.0* 12.5 12.0 reduction of acute phase reactants. Moreover, several clinical measures continued Total cholesterol (mmol/l) 4.8 3.5* 3.6* 5.0** 5.0** 2 to improve during the follow up phase, suggesting that rosuvastatin has longlasting HDL-cholesterol (mmol/l) 1.3 1.3 1.5 1.3 1.2 beneficial effects. Accumulating evidence suggests that statins exert anti- LDL-cholesterol (mmol/l) 2.8 1.1* 1.8* 3.3** 3.2** inflammatory properties through modulation of the immune response. The immune Triglycerides (mmol/l) 1.4 1.2 1.3 1.5 1.8 response is up regulated in the inflammatory disorder AS and we suggested that 3 BASDAI 6.3 5.6 5.8 5.2 5.6 suppression of the immune response by statins would lead to clinical improvement BASFI 4.7 5.4 4.9 5.0 4.9 accompanied by lower levels of inflammation markers, such as ESR and CRP. The BASMI 4.4 4.0 3.8 3.6 4.2 results of our investigation are in line with this hypothesis, but clearly, confirmatory 4 randomised studies are required. General wellbeing according to doctor 4.4 4.5 3.9 4.2 4.0 Finally, as AS is associated with an increased cardiovascular risk [5], the use of General wellbeing according to patient 6.4 5.7 5.3 5.6 5.7 statins might ultimately lead to reduction of this risk by two pathways: one through Pain score 6.5 6.8 6.2 5.7 5.2 the lipoprotein metabolism and the other by beneficial effects on the underlying *Significant changes compared with baseline (Wilcoxon’s signed ranks test), p 0.05; 5 inflammatory process in AS. **significant changes compared with 12 weeks (Wilcoxon’s signed rank test), p 0.05. 6

7

REFERENCES

[1] Scalapino KJ, Davis JC Jr. The treatment of ankylosing spondylitis. Clin Exp 8 Med 2003;4:159–65. [2] Braun J, Sieper J. Biological therapies in the spondylarthritides—the current state. Rheumatology (Oxford) 2004;43:1072–84. 9 [3] Halcox JPJ, Deanfield JE. Beyond the laboratory: clinical implications for statin pleitropy. Circulation 2004;109(suppl 1):II42–8. [4] McCarey DW, McInnes IB, Madhok R, Hampson R, Scherbakova O, Ford I, et al. Trial of atorvastatin in rheumatoid arthritis (TARA): double-blind, randomised 10 placebo-controlled trial. Lancet 2004;363:2015–21. [5] Peters MJ, Van der Horst-Bruinsma IE, Dijkmans BA, Nurmohamed MT. Cardiovascular risk profile of patients with spondylarthropathies, particularly ankylosing spondylitis and psoriatic arthritis. Semin Arthritis Rheum 2004;34:585–92.

54 55 CHAPTER 4

Double blind, randomised, placebo controlled study of leflunomide in the treatment of active ankylosing spondylitis.

J C van Denderen1 M van der Paardt1 M T Nurmohamed1 Y M M A de Ryck3 B A C Dijkmans2 I E van der Horst-Bruinsma2

Annals of the Rheumatic Diseases 2005;64:1761-1764.

1 Department of Rheumatology, Jan van Breemen Instituut, Amsterdam. 2 Department of Rheumatology, VU University Medical Center, Amsterdam. 3 Aventis Pharma, Hoevelaken. Chapter 4 Leflunomide in AS

ABSTRACT INTRODUCTION

Objective Ankylosing spondylitis (AS) is an characterised by chronic To assess the efficacy and safety of leflunomide in active ankylosing spondylitis (AS) inflammation of the sacroiliac and spinal joints and entheses. The disease occurs compared with placebo in a 24 week pilot study. mainly in young adults and can lead to stiffness and deformity of the vertebral 1 column, with invalidating deformities. AS is often accompanied by extraspinal Methods manifestations as arthritis of the peripheral joints, and involvement of the eye (acute In a single centre randomised, double blind, placebo controlled study, 45 patients anterior uveitis), heart, and . with active AS were randomised to either leflunomide 20 mg daily (n = 30) or placebo 2 (n = 15). Active disease was defined as a score of at least 4 on the Bath ankylosing Treatment was, until recently, mainly based on non steroidal anti-inflammatory spondylitis disease activity index (0–10), and pain of at least 4 on a visual analogue drugs (NSAIDs) and physical therapy [1]. The disease modifying antirheumatic scale (0–10). The primary efficacy variable at week 24 was the 20% response rate, drugs (DMARDs), for example sulfasalazine and methotrexate, seem to be less 3 as recommended by the Assessments in Ankylosing Spondylitis (ASAS) working beneficial in AS than in other rheumatic diseases such as rheumatoid arthritis [2]. group. Secondary outcome variables included general wellbeing, metrology index, Recent studies have shown that the tumour necrosis factor α (TNFα) blocking agents swollen joint count, erythrocyte sedimentation rate, and C reactive protein. infliximab and etanercept are very effective in a large proportion of patients with AS [3, 4]. These powerful drugs, however, are costly and are sometimes accompanied 4 Results by severe adverse effects such as opportunistic infections. Moreover, these agents In all, 13 women and 32 men were studied. Demographic and disease indices were fail to reach efficacy in approximately 30% of the patients [5]. For these reasons comparable between the two treatment groups at baseline. The rate of ASAS 20% we investigated the efficacy and safety of another DMARD, leflunomide: a drug responders was not significantly different: 27% in the leflunomide treated patients proven to be effective in rheumatoid arthritis [6–9]. In addition, leflunomide shows 5 and 20% in the placebo group (95% confidence interval, −32% to 19%). No significant beneficial effects in patients with psoriatic arthritis which, like AS, belongs to the differences were found between the treatment groups in mean changes of the group of spondyloarthritides [10–13]. In the present study the efficacy and safety secondary outcome variables. Eleven patients were withdrawn prematurely from of leflunomide was investigated in patients with active AS in a randomised, double 6 the study because of adverse events (7), lack of efficacy (3), and non-compliance (1). blind, placebo controlled trial. Most frequently adverse events were gastrointestinal side effects and skin disorders. METHODS Conclusion 7 In this placebo controlled study, leflunomide treatment did not result in a significant Consecutive patients with AS were recruited from the outpatient rheumatology improvement of the ASAS 20% response in active ankylosing spondylitis. No clinic of the Jan van Breemen Institute (referral centre) and the surrounding unexpected or severe adverse events occurred. hospitals in Amsterdam. The study was carried out in the period from March 2002 to 8 September 2003. The study group comprised 45 patients aged 18 to 70 years with active definite AS, diagnosed according to the modified New York criteria [14]. Active disease was defined as: at least a 4 point score on the Bath ankylosing spondylitis disease activity index (BASDAI; scale 0–10) and a pain score of at least 4 on a visual 9 analogue scale (VAS; scale 0–10) at screening.

NSAIDs and corticosteroids up to a maximum daily dose of 15 mg prednisone 10 were allowed on an unchanged regimen for at least 30 days before the study drug administration and throughout the study. The use of DMARDs (particularly sulfasalazine, methotrexate, and TNF blocking agents) during the study and within the 30 days before the study was not permitted.

58 59 Chapter 4 Leflunomide in AS

Both men and women were required to practice contraception. In female patients of The primary efficacy variable was the 20% response rate, as recommended by the childbearing potential a urine pregnancy test was done at baseline and any pregnant Assessments in Ankylosing Spondylitis (ASAS) working Group [19]. Each patient women were excluded. was classified as a treatment responder or non-responder at the end point (week 24 or last available observation under treatment). The ASAS 20% response was defined Other exclusion criteria were: impaired hepatic enzyme tests (alanine and aspartate as improvement of ≥20% and absolute improvement of ≥1 unit (on a scale of 0 to 10) 1 transaminases more than twice the upper laboratory limit of normal), impaired in at least three of the following four domains and absence of deterioration (change renal function (serum creatinine >110 μmol/l), untreated hypertension, malignancy, for the worse of ≥20% and net worsening of ≥1 unit) in the remaining domain: diagnosis of other inflammatory joint diseases, impaired bone marrow function, patient global assessment, pain (VAS), BASFI, and inflammation (the mean of the recent serious infections, and any clinically relevant cardiovascular, hepatic, two morning stiffness related BASDAI VAS scores). 2 neurological, endocrine. or other major systemic diseases. The responder rates in the two treatment groups were compared using the binomial Design test for two samples, and the 95% confidence intervals were calculated. Forty five eligible patients were randomised at baseline (within four weeks after Subgroup analyses were done in patients with an ESR of >30 mm/h or a C reactive 3 screening) to receive either leflunomide (n = 30) or matching placebo (n = 15) for 24 protein of >10 mg/l, or both, at baseline, and in patients with peripheral arthritis weeks. The dosage schedule included a loading dose of 100 mg (given on day 1, (SJC >1 at baseline). 8, and 15) and leflunomide 20 mg on the other days during the first three weeks, 4 followed by a maintenance dose of 20 mg daily. We often apply this slightly modified The secondary efficacy variables at end point were compared with the assessments dosing schedule, as we have the impression that the rate of gastrointestinal at baseline and 95% confidence intervals for the differences between leflunomide complaints is less than with the standard schedule. Compliance was assessed by and placebo for the mean changes from baseline were calculated. Student’s t test tablet counts of returned study preparations and was calculated on the basis of the was used for variables with a normal distribution and the Mann–Whitney U test for 5 number of days in the study. variables with a non-normal distribution. A two sided p value of less than 0.05 was considered significant. Efficacy and safety indices (vital signs and adverse events) were assessed at 6 baseline, week 4, week 12, and week 24. The following efficacy variables were used: The primary and secondary efficacy analyses and safety analysis were carried out BASDAI [15], pain assessed on a VAS (0–10), BASFI [16], Bath ankylosing spondylitis on an intention to treat basis. global score (BASG) [17], Bath ankylosing spondylitis metrology index (BASMI) [18], 44 swollen joint count (SJC), general wellbeing on a VAS (0–10) according to the The efficacy analysis was also done for the per-protocol (PP) population. We defined 7 physician, erythrocyte sedimentation rate (ESR), and C reactive protein. the PP population as all patients treated with a minimum drug exposure of eight weeks, and excluding one patient in whom the prednisone dose was increased. Blood tests (haematology and blood chemistry) were monitored every two weeks The PP population (36 patients) was identified before unblinding the treatment 8 during the first six weeks and at regular intervals later during the study period. allocation. The study was approved by the local ethics review board, and written informed Statistics consent was obtained from all patients. Based on an assumed 50% response rate for leflunomide and 20% for placebo, 9 adjusting for a non-evaluable rate of 25%, 61 patients per treatment would RESULTS have been needed to reach a level of significance of 0.05 with a power of 80%. Underlining the explorative character of this trial, and because of obvious feasibility Baseline characteristics 10 issues in this single centre study, it was decided to randomise 45 patients. Expecting The demographic data of the patients and disease characteristics at baseline are a higher dropout rate in the leflunomide group, a randomisation ratio was defined summarised in table 1. The variables were comparable and not statistically different with a 2:1 balance (30 leflunomide patients and 15 placebo patients) so as to ensure between the two treatment groups. Fourteen patients had a history of uveitis (eight a minimum collection of efficacy and safety data in a new indication for leflunomide. leflunomide and six placebo). Two patients in the leflunomide group and one in the Altogether, the power of the study was reduced to 65%, which reasonably can be placebo group had inactive inflammatory bowel disease and one patient in both said to show at most an effect trend.

60 61 Chapter 4 Leflunomide in AS

groups had psoriasis. NSAIDs were used by 93% of the leflunomide and 87% of (BASMI), swollen joint count or C reactive protein between the treatment groups the placebo treated patients. Corticosteroids were used by one patient. None of the (table 2). The mean change in ESR was the only variable that differed significantly patients had ever used a TNF blocking agent. between the treatment groups in favour of the placebo group. This observation was the result of a skewed distribution of the ESR, because the differences between the median changes were much smaller. 1 Table 1. Demographic and mean (or median) baseline characteristics (and ranges) of 45 patients with active ankylosing spondylitis.

Characteristic Placebo (n = 15) Leflunomide (n = 30) Table 2. Mean efficacy variables at baseline and the mean change after 24 weeks with the 95% confidence interval for the difference in change between the treatment Male sex (%) 10 (67) 22 (73) 2 groups. % HLA-B27 positive 100 (14/14) 81 (21/26) Age (years) 39 (25 to 58) 42 (22 to 66) Placebo (n = 15) Leflunomide (n = 30) Median disease duration (years) 5.7 (0.3 to 14.1) 9.4 (0.3 to 38.6) Mean Mean 3 Variable Baseline change Baseline change 95% CI* p Value Peripheral arthritis (%)* 2(13) 2 (7) BASG last week (0–10) 5.8 −0.6 6.9 −1.3 (−0.9 to 2.4) 0.32 BASDAI (0–10) 5.9 (2.9 to 9.7) 6.4 (2.7 to 9.5) BASG last six months 6.7 −1.5 7. 4 −0.7 (−2.2 to 0.6) 0.26 ESR (mm/h) 18 (2 to 54) 22 (2 to 66) Pain (0–10) 6.2 −0.5 6.8 −1.4 (−0.9 to 2.8) 0.28 4 *Number of patients with at least one swollen joint. BASFI (0–10) 5.4 −0.4 5.2 0.0 (−1.3 to 0.5) 0.35 BASDAI, Bath ankylosing spondylitis disease activity index; ESR, erythrocyte sedimentation BASDAI (0–10) 5.9 −0.3 6.4 −1.1 (−0.5 to 2.0) 0.20 rate. BASMI (0–10) 3.3 0.3 4.2 0.0 (−0.1 to 0.8) 0.10 5 SJC (0–44) 0.4 −0.2 0.1 0.2 (−0.9 to 0.1) 0.09 Primary efficacy variable Global physician (0–10)† 5.1 −0.7 5.5 −0.5 (−1.1 to 0.8) 0.70 The number of responders according to the ASAS 20% definition was 11: 8 (27%) in ESR (mm/h) 18.1 −1.9 22.1 8.4 (−17.8 to −2.9) 0.002 6 the leflunomide group and 3 (20%) in the placebo group (95% confidence interval C reactive protein (mg/l) 21.4 −6.1 27.0 6.5 (−30.9 to 5.8) 0.09

(CI) for difference, −32% to 19%). *95% confidence interval for difference of mean change between placebo and leflunomide. †General wellbeing of the patient according to physician. In the subgroup with an ESR of ≥30 mm/h, or a C reactive protein of ≥10 mg/l, or BASDAI, Bath ankylosing spondylitis disease activity index; BASFI, Bath ankylosing spondylitis 7 both, at baseline (n = 26), the response rate was 32% and 29% for the leflunomide functional index; BASG, Bath ankylosing spondylitis global score; BASMI, Bath ankylosing and the placebo group, respectively (−42% to 36%). spondylitis metrology index; ESR, erythrocyte sedimentation rate; SJC, swollen joint count. 8 In the subgroup of patients with peripheral arthritis (SJS ≥1 at baseline), which included only four patients, the response rate was 50% in the leflunomide and 100% Adverse events and dropouts in the placebo group (−19% to 100%). Adverse events were seen in 24 leflunomide cases (80%) and 10 placebo cases (67%). The adverse events were classified as possibly drug related in 18 (60%) and 9 In the per-protocol population the response rate was 6/24 (25%) in the patients six (40%) of the leflunomide and placebo treated patients, respectively. No serious treated with leflunomide and 2/12 (17%) in the placebo treated patients (−36% to adverse events were seen. Most frequent or clinically relevant adverse events were: 19%). gastrointestinal disorders (57% in the leflunomide and 33% in the placebo treated 10 patients, respectively); upper respiratory tract infections (17% in the leflunomide and Secondary efficacy variables 27% in the placebo treated patients); dermatitis and prurigo (13% in both groups); The mean secondary efficacy variables at baseline and the mean changes after 24 fatigue (in 13% of the placebo treated patients); deep venous thrombosis (in one weeks did not show any significant differences of global disease activity (BASG), placebo patient). In the leflunomide group, a rise in liver enzymes to three times the disease activity index (BASDAI), functional index (BASFI), pain, metrology index upper normal limit was observed in one patient with borderline liver enzyme tests at

62 63 Chapter 4 Leflunomide in AS

baseline. No clinically relevant changes were found in the other laboratory variables. Only one other study has investigated the efficacy of leflunomide in AS; this was No changes were seen in the mean blood pressures. Antihypertensive treatment had an open label non-comparative trial of 20 AS patients who were treated for 24 to be increased in one patient treated with leflunomide. weeks [20]. Ten patients were prematurely withdrawn because of lack of efficacy, side effects, or non-compliance. In line with our results, leflunomide seemed to Eleven patients were withdrawn prematurely from the study (fig 1), for the following be ineffective. The only significant improvement observed was in a subgroup 1 reasons: adverse events (five leflunomide patients, two placebo patients); lack of of 10 patients suffering from peripheral arthritis. In our study, only four patients efficacy (three leflunomide patients); non-compliance (one placebo patient). Four had peripheral arthritis in at least one joint and therefore we could not draw any of these dropouts were responders (three leflunomide patients and one placebo conclusions about the efficacy of leflunomide in this subgroup. patient). Adverse events leading to withdrawal were gastrointestinal disorders (4), 2 malaise (1), exacerbation of pain (1), headache (1), and erectile dysfunction (1). The efficacy of leflunomide in AS in our study appears disappointing but it is consistent with the experience with other DMARDs in AS. Most studies with DMARDs in AS, particularly sulfasalazine, show improvement of the peripheral arthritis and ESR, but no significant improvement in the axial complaints [2, 21]. 3 Mesalazine, which is very effective in inflammatory bowel disease, was not shown to be effective in AS [22]. Limited data are available of methotrexate in AS, but a recent small placebo controlled trial (in which a lower dose was used than in most 4 studies in rheumatoid arthritis) showed favourable results [23]. The large number of patients with peripheral arthritis in that study (65% of the 17 recruited patients) was striking. 5 In rheumatoid arthritis leflunomide reduces the numbers of macrophages and T cells infiltrating the synovium, and the expression of adhesion molecules, proinflammatory cytokines, and mediators of joint destruction such as matrix 6 metalloproteinases (MMPs) [24]. Whereas the molecular pathways have been thoroughly analysed in rheumatoid arthritis, the mechanisms of joint inflammation Figure 1. Trial profile. Five screened patients are not included because they did not meet the and destruction are less well understood in AS. In sacroiliac biopsies from 32 criteria for disease activity, and two patients because they wanted to have a child. AE, adverse patients with spondyloarthritis, T cells and macrophages were the predominant 7 events. cells, suggesting an important role for these cell types in the pathogenesis [25]. Moreover, TNFα blocking treatment in AS downregulates both cytokines secreted by T cells and MMPs in the synovium in spondyloarthritis [26, 27]. Therefore an effect Compliance 8 of leflunomide on AS could be expected. It is interesting that significantly higher On the basis of the number of days in the study (leflunomide, mean 134 days; serum MMP-3 concentrations are encountered in AS patients with peripheral joint placebo, mean 142 days) and the number of returned tablets, the calculated disease than in those with only axial symptoms [27]. This latter finding could be the compliance was 96% in the leflunomide group and 93% in the placebo group. pathophysiological explanation for a possible effect of leflunomide in cases of AS 9 with peripheral joint disease. DISCUSSION

It is clear that leflunomide is not effective when compared with the very efficacious Leflunomide did not result in significant clinical improvement in active ankylosing 10 TNF-α blocking agents, although direct comparative trials have not been conducted. spondylitis in this placebo controlled study. The number of AS patients who fulfilled An interesting observation from a trial of anti-TNFα treatment in AS was that the ASAS 20% response rate was higher in the leflunomide group than in the patients with a raised C reactive protein had greater benefit from the treatment than placebo group, but the difference did not reach statistical significance. The changes those with lower levels [3]. However, in our study no difference in responder rate in the individual secondary disease outcome indices were comparable. Importantly, between the treatment groups was seen in the subgroup with an ESR of ≥30 mm/ no serious or unexpected adverse events were encountered. hour or a C reactive protein of ≥10 mg/l, or both, at baseline.

64 65 Chapter 4 Leflunomide in AS

Conclusion REFERENCES This is the first double blind, placebo controlled study of leflunomide in active AS. Unfortunately, it was not possible to show that it had significantly better efficacy [1] Van der Linden S, Van Tubergen A, Hidding A. Physiotherapy in ankylosing spondylitis: than placebo. Whether or not this drug is beneficial for AS patients with peripheral what is the evidence? Clin Exp Rheumatol 2002;20 (suppl 28) :S60–4. arthritis remains to be established, as in our study the number of such patients was [2] Van der Horst-Bruinsma IE, Clegg DO, Dijkmans BA. Treatment of ankylosing spondylitis 1 too small to allow us to draw meaningful conclusions. This might be an interesting with disease modifying antirheumatic drugs. Clin Exp Rheumatol 2002;20 (suppl 28) :S67–70. subject for future research. [3] Braun J, Brandt J, Listing J, Zink A, Alten R, Golder W, et al. Treatment of active ankylosing spondylitis with infliximab: a randomised controlled multicentre trial. Lancet 2002;359:1187–93. 2 [4] Gorman JD, Sack KE, Davis JC. Treatment of ankylosing spondylitis by inhibition of tumor necrosis factor α. N Engl J Med 2002;346:1349–56. [5] Braun J, Brandt J, Listing J, Zink A, Alten R, Burmester G. Long-term efficacy and 3 safety of infliximab in the treatment of ankylosing spondylitis. An open, observational, extension study of a three-month, randomised, placebo-controlled trial. Arthritis Rheum 2003;48:2224–33. [6] Mladenovic V, Domljan Z, Rozman B, Jajic I, Mihajlovic D, Dordevic J, et al. Safety and 4 effectiveness of leflunomide in the treatment of patients with active rheumatoid arthritis: results of a randomised, placebo-controlled, phase II study. Arthritis Rheum 1995;38:1595– 603. [7] Smolen JS, Kalden JR, Scott DL, Rozman B, Kvien TK, Larsen L, et al. Efficacy and safety 5 of leflunomide compared with placebo and sulphasalazine in active rheumatoid arthritis: a double-blind, randomised, multicenter trial. Lancet 1999;353:259–66. [8] Strand V, Cohen S, Schiff M, Weaver A, Fleischmann R, Cannon G, et al. Treatment of active rheumatoid arthritis with leflunomide compared with placebo and methotrexate. Arch 6 Intern Med 1999;159:2542–50. [9] Scott DL, Smolen JS, Kalden JR, van de Putte LB, Larsen A, Kvien TK, for the European Leflunomide Study Group, et al. Treatment of active rheumatoid arthritis with leflunomide: two year follow up of a double blind, placebo controlled trial versus sulfasalazine. Ann 7 Rheum Dis 2001;60:913–23. [10] Kaltwasser JP, Nash P, Gladman D, Rosen CF, Behrens F, Jones P, for the Treatment of Psoriatic Arthritis Study Group, et al. Efficacy and safety of Leflunomide in the treatment of psoriatic arthritis and psoriasis. A multinational, double-blind, randomised, placebo- 8 controlled clinical trial. Arthritis Rheum 2004;50:1939–50. [11] Scali JJ, Visentini S, Salomon G, Castelli G, Barcena P, Morales E. Treatment of psoriatic arthritis: open prospective study comparing efficacy and safety profile between leflunomide vs methotrexate. Ann Rheum Dis 2000;59 (suppl 1):POS-566. 9 [12] Liang G, Barr W. Long term follow-up of the use of leflunomide in recalcitrant psoriatic arthritis and psoriasis. Arthritis Rheum 2001;44 (suppl):POS-413. [13] Cuchacovich M, Soto L. Leflunomide decreases joint erosions and induces reparative changes in a patient with psoriatic arthritis. Ann Rheum Dis 2002;61:942–3. 10 [14] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8.

66 67 Chapter 4 Leflunomide in AS

[15] Garret S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994;21:2286–91. [16] Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mallorie P, et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index. J Rheumatol 1994;21:2281–5. 1 [17] Jones SD, Steiner A, Garrett SL, Calin A. The Bath Ankylosing Patient Global score (BAS-G). Br J Rheumatol 1996;35:66–77. [18] Jones SD, Porter J, Garrett SL, Kennedy LG, Whitelock H, Calin A. A new scoring system for the Bath Ankylosing Spondylitis Metrology Index (BASMI). J Rheumatol 1995;22:1609. 2 [19] Anderson JJ, Baron G, van der Heijde D, Felson DT, Dougados M. Ankylosing spondylitis assessment group preliminary definition of short-term improvement in ankylosing spondylitis. Arthritis Rheum 2001;44:1876–86. 3 [20] Haibel H, Rudwaleit M, Braun J, Sieper J. Six month open label trial of leflunomide in active ankylosing spondylitis. Ann Rheum Dis 2005;64:124–6. [21] Clegg DO, Reda DJ, Weisman MH, Blackburn WD, Cush JJ, Cannon GW, et al. Comparison of sulfasalazine and placebo in the treatment of ankylosing spondylitis; a Department of 4 Veterans Affairs cooperative study. Arthritis Rheum 1996;39:2004–12. [22] Van Denderen JC, van der Horst-Bruinsma IE, Bezemer PD, Dijkmans BA. Efficacy and safety of mesalazine (Salofalk®) in an open study of 20 patients with ankylosing spondylitis. J Rheumatol 2003;30:1558–60. 5 [23] Gonzalez-Lopez L, Garcia-Conzalez A, Vasques-Del-Mercado M, Munoz-Valle JF, Gamez- Nava JL. Efficacy of methotrexate in ankylosing spondylitis: a randomized, double blind, placebo controlled trial. J Rheumatol 2004;31:1568–74. [24] Kraan MC, Reece RJ, Barg EC, Smeets TJ, Farnell J, Rosenburg R, et al. Modulation of 6 inflammation and metalloproteinase expression in synovial tissue by leflunomide and methotrexate in patients with active rheumatoid arthritis: findings in a prospective, randomised, double-blind, parallel-design clinical trial in thirty-nine patients at two centers. Arthritis Rheum 2000;43:1820–30. 7 [25] Bollow M, Fischer T, Reisshauer H, Backhaus M, Sieper J, Hamm B, et al. Quantitative analyses of sacroiliac biopsies in spondyloarthropathies: T cells and macrophages predominate in early and active sacroiliitis – cellularity correlates with the degree of enhancement detected by magnetic resonance imaging. Ann Rheum Dis 2000;59:135–40. 8 [26] Zou J, Rudwaleit M, Brandt J, Thiel A, Braun J, Sieper J. Down-regulation of the nonspecific and antigen-specific T cell cytokine response in ankylosing spondylitis during treatment with infliximab. Arthritis Rheum 2003;48:780–90. [27] Vandooren B, Kruithof E, Yu DT, Rihl M, Gu J, De Rycke L, et al. Involvement of matrix 9 metalloproteinases and their inhibitors in peripheral synovitis and down-regulation by tumor necrosis factor α blockade in spondylarthropathy. Arthritis Rheum 2004;50:2942–53. 10

68 69 CHAPTER 5

Decreased clinical response to infliximab in ankylosing spondylitis is correlated with anti-infliximab formation.

M K de Vries1 G J Wolbink2 S O Stapel2 H de Vrieze2 J C van Denderen3 B A C Dijkmans1 L A Aarden2 I E van der Horst-Bruinsma1

Annals of the Rheumatic Diseases 2007;66:1252-1254.

1 VU University Medical Center, Amsterdam. 2 Sanquin Research, Amsterdam. 3 Jan van Breemen Instituut, Amsterdam. Chapter 5 Anti-infliximab

ABSTRACT INTRODUCTION

Objective Large randomised clinical trials have shown that tumour necrosis factor blocking Correlation of serum trough infliximab levels and antibodies to infliximab (anti- agents such as infliximab are very effective in ankylosing spondylitis [1]. It is infliximab) with clinical response in ankylosing spondylitis. unknown why >30% of patients with ankylosing spondylitis fail to respond, or why 1 some initial responders lose responsiveness during treatment and in some cases Methods even develop an infusion reaction. In accordance with the international ASsessment in Ankylosing Spondylitis (ASAS) consensus statement, patients were treated with infliximab (5 mg/kg) every 6 weeks The non-responsiveness to infliximab might be due to the development of 2 after a starting regimen. Preinfusion sera were collected at baseline, 24 and 54 antibodies against it, which has been described in patients with rheumatoid arthritis weeks. At every visit, the 20% improvement response (ASAS-20) was assessed and and Crohn’s disease [2–5]. In ankylosing spondylitis, we recently showed in a small laboratory tests performed. group of patients that detection of anti-infliximab was associated with undetectable 3 serum trough infliximab levels, a reduced response to treatment and a higher risk of Results infusion reactions [6]. 24 of the 38 (63%) patients fulfilled ASAS-20 response criteria after 24 weeks of treatment and 21 (53%) after 54 weeks. After 54 weeks, 11 (29%) patients showed The aim of this study was to evaluate these data in a larger group of patients with 4 undetectable serum trough infliximab levels and detectable anti-infliximab; ankylosing spondylitis who were treated for a longer period of time and to specify six of these patients developed an infusion reaction. Anti-infliximab was found the influence on infliximab levels. significantly more often (p = 0.04) in ASAS-20 non-responders compared with responders at week 54. Serum trough infliximab levels were significantly (p<0.0001) METHODS 5 lower in patients with (mean 0.02 mg/l) than in those without (12.7 mg/l) anti- infliximab. All consecutive patients with ankylosing spondylitis (according to the 1984 modified New York Criteria [7]) who received treatment with infliximab in our centre were 6 Conclusion included in this study. In ankylosing spondylitis, high levels of serum trough infliximab correlated with a good clinical response. Detection of anti-infliximab within 54 weeks is associated Disease activity was measured with the Bath Ankylosing Spondylitis Disease Activity with undetectable serum trough infliximab levels, reduced response to treatment Index (BASDAI) [8] and the ASsessment in Ankylosing Spondylitis 20% response 7 and increased risk of developing an infusion reaction. criteria (ASAS-20) [9]. Active disease was defined as a BASDAI score ≥4. Response to treatment with infliximab was defined as fulfilment of the ASAS-20 response criteria. 8 Patients with ankylosing spondylitis were treated with intravenous infliximab, 5 mg/kg bodyweight at baseline, weeks 2 and 6, and every 6 weeks thereafter. This treatment was initiated in accordance with the international ASAS consensus statement [9]. In case of decrease of clinical response, the dose of infliximab was 9 increased to 7.5 mg/kg.

At each visit, the presence of infections, side-effects or infusion reactions, and the 10 cause for discontinuation of therapy were recorded. Questionnaires and routine laboratory tests were obtained. Preinfusion sera were collected at baseline, weeks 24 and 54, before any dose escalation and at two consecutive visits after dose escalation. After 24 weeks of treatment, serum samples were collected from 15 patients to measure infliximab levels 2 weeks after the infliximab infusion.

72 73 Chapter 5 Anti-infliximab

Validated immunoassays (Sanquin Research, Amsterdam, the Netherlands) were Table 1. Demographic and clinical variables at baseline, week 24 and at week 54 of used for detection of anti-infliximab and serum trough infliximab levels [5]. Trough patients with ankylosing spondylitis (n = 38). serum infliximab levels were measured by ELISA, based on the principle that Variables Baseline Week 24 Week 54 infliximab is captured through its ability to bind tumour necrosis factor-α. The assay, Male gender (%) 26 (68) — — which was described previously, was modified recently. It currently uses specific Age (years) 40 (10) — — 1 polyclonal rabbit antibodies to infliximab for detection instead of the monoclonal HLA B27+ (%) 32 (84) — — anti human IgG that was previously used. The sensitivity of detection is 0.0003 mg/l. IBD (%) 6 (16) — — Use of corticosteroids (%) 3 (8) — — A radioimmunoassay was used for anti-infliximab detection [5]. Arbitrary units per 2 Use of other immunosuppressives (%) 6 (16) — — ml (AU/ml) were expressed as absolute amounts of infliximab-specific IgG (mg/l) BASDAI 6.4 (1.2) 3.6 (2.6)‡ 4.1 (3.0‡ (1 AU = 12 ng of infliximab-specific IgG) [10]. The cut-off value for IgG anti-infliximab was determined by assaying in our anti-infliximab test 100 plasma samples from Morning stiffness† 6.3 (2.2) 3.0 (2.5)‡ 3.5 (3.2)‡ blood donors sent to Sanquin for IgG anti-tetanus toxoid testing. The average result GDA 6.8 (1.3) 4.3 (2.9)‡ 4.9 (3.4)§ 3 (AU/ml) + 6 SD was 12 AU/ml (0.144 mg/l). C-reactive protein (normal<8.0 mg/l) 37 (34.2) 9.3 (10.7)‡ 15.8 (21.1)¶ Detectable serum trough infliximab (%) 0 31 (82) 27 (71) The clinical data and presence of HLA B27 were used to correlate disease activity Anti-infliximab (%) 0 7 (18) 11 (29) 4 with serum trough infliximab levels and anti-infliximab levels. Differences between BASDAI, Bath Ankylosing Spondylitis Disease Activity Index (0–10 cm); GDA VAS, Global Dis- groups were tested with the Mann–Whitney U test. Associations were calculated ease Activity Visual Analogue Scale (0–10 cm); HLA B27, human lymphocyte antigen B27; IBD, with logistic regression. The threshold for significance was set at p<0.05. The last Except when indicated otherwise, the values are the mean (SD). inflammatory bowel disease. observation was carried forward for patients who dropped out before week 54. † Mean of item 5+6 of the BASDAI (0–10 cm). 5 Compared with baseline: ‡ p<0.001; § p = 0.005; ¶ p = 0.003. RESULTS

In total, 9% (1 of 11) patients with detectable anti-infliximab was classified as a 6 Demographic and clinical characteristics of the 38 patients included are shown in responder at week 54, compared with 74% (20 of 27) of patients without anti- table 1. Four patients were lost to follow-up before week 54: one wanted to become infliximab (figure 2). pregnant, one preferred to be treated in a hospital nearby and two because of comorbidities. 7

There was a significant decrease in BASDAI, morning stiffness, global disease activity and C-reactive protein after 24 and 54 weeks of treatment (table 1) and 8 all pre-treatment samples showed undetectable infliximab levels and no anti- infliximab. We did not detect anti-infliximab in the presence of infliximab.

After 24 weeks, 24 patients (63%) met ASAS-20 response criteria. Responders 9 showed higher mean serum trough infliximab levels, and only two patients (8%) showed anti-infliximab, compared with 5 (36%) of the non-responders (p = 0.08). 10 After 54 weeks of treatment, ASAS-20 response criteria were met by 21 patients (53%). The mean serum trough infliximab level for responders was significantly (p<0.01) higher that that of the non-responders (8.2 mg/l vs 6.3 mg/l; figure 1) and anti-infliximab was significantly (p<0.04) more often found in non-responders. Only Figure 1. Serum trough infliximab level for responders (n = 21; 8.2 mg/l) and non-responders 5% (1 of 21) of the responders showed anti-infliximab, compared with 59% (10 of 17) (n = 17; 6.3 mg/l) according to the ASAS-20 response criteria, at week 54 (p = 0.018). of the non-responders.

74 75 Chapter 5 Anti-infliximab

DISCUSSION

A good clinical response of ankylosing spondylitis to treatment with infliximab was correlated with the presence of high serum trough infliximab levels and the absence of anti-infliximab antibodies, and inefficacy with the reverse. Moreover, these data 1 demonstrate that anti-infliximab antibodies precede an infusion reaction.

The mechanism of the decrease in efficacy can be explained by the lower serum trough infliximab levels, probably caused by enhanced clearance due to immune 2 Figure 2. Percentage of patients (n = 38) with (9%) and without (74%) anti-infliximab fulfilling complex formation of anti-infliximab antibodies and infliximab. A recent study ASAS-20 response criteria at week 54 (p<0.001). in RA showed an enhanced clearance as a consequence of this process and an accumulation in the macrophage–phagocyte system (liver and spleen) [11]. 3 Indeed, in those patients with ankylosing spondylitis who developed detectable After correction for probable confounding variables such as gender and human anti-infliximab within 54 weeks of treatment with infliximab, a significantly lower leucocyte antigen B27 (HLA-B27), the absence of anti-infliximab remained a infliximab level was found 2 weeks after infusion compared with patients who did significant determinant for ASAS-20 response, with an odds ratio (OR) of 100 not develop anti-infliximab. (95% CI 5.2 to 1000). Remarkably, the presence of anti-infliximab was significantly 4 associated with the absence of HLA-B27 (OR = 7.1; 95% CI 1.1 to 47.6; Pearson chi- Often, the infliximab dose is increased in ankylosing spondylitis when squared test, p = 0.03). responsiveness decreases, but reasons for dose escalation in ankylosing spondylitis are not yet well defined. In our small sample, no clear increase in serum trough 5 Two weeks after the infusion of week 24, significantly lower infliximab levels were infliximab level after dose escalation was shown. measured (20 mg/l compared with 51 mg/l; p<0.01) in patients who developed anti- infliximab within 54 weeks of treatment. Another option is to try to prevent anti-infliximab formation with the concomitant 6 administration of other immunosuppressive drugs such as methotrexate; however, In 12 patients, dose was increased within the 54 weeks, because of insufficient this medication is not efficacious in ankylosing spondylitis [12]. clinical response. Nine (75%) of these patients showed anti-infliximab antibodies. Increase in dose did not result in a significant increase of the serum trough Remarkably, absence of HLA B27 shows significant correlation with anti-infliximab 7 infliximab level (p = 0.33), or a significant decrease in the anti-infliximab level formation. Further genetic evaluation will be performed to unravel this interesting (p = 0.90) and BASDAI (p = 0.39). However, 2 of 12 patients reported longer duration of observation. effect. 8 It also has to be investigated whether coadministration of immunosuppressive Infusion reactions occurred in six patients. Most reactions were mild, and all patients drugs inhibits anti-infliximab formation, and whether infliximab levels can be used recovered after supportive therapy. Treatment with infliximab was stopped in each for determination of the optimum dose of infliximab in ankylosing spondylitis. case. Every infusion reaction was preceded by development of anti-infliximab 9 and consequently undetectable serum trough infliximab levels. All antibodies to In accordance with our previous report, the efficacy of infliximab in ankylosing infliximab consisted of IgG1 and IgG4 subtypes. Although these infusion reactions spondylitis is clearly related to infliximab levels and the formation of anti-infliximab resemble a type 1 allergic reaction, no IgE was detected. One patient’s pre-infusion antibodies. Detection of anti-infliximab antibodies within 54 weeks is associated serum contained an anti-infliximab level of 6.4 g/l, indicating that approximately half 10 with undetectable serum trough infliximab levels, reduced response to treatment of his total serum IgG consisted of infliximab-specific antibodies. and increased risk of development of an infusion reaction.

76 77 Chapter 5 Anti-infliximab

REFERENCES

[1] Van der Heijde D. Dijkmans B, Geusens P, Sieper J, DeWoody K, Williamson P, et al. Efficacy and safety of infliximab in patients with ankylosing spondylitis: results of a randomized, placebo-controlled trial (ASSERT), Arthritis Rheum 2005;52:582–91. [2] Baert F, Noman M, Vermeire S, Van AG, D’ HG, Carbonez A, et al. Influence of 1 immunogenicity on the long-term efficacy of infliximab in Crohn’s disease. N Engl J Med 2003;348:601–8. [3] Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2 2002;359:1541–9. [4] St Clair EW, Wagner CL, Fasanmade AA, Wang B, Schaible T, Kavanaugh A, et al. The relationship of serum infliximab concentrations to clinical improvement in rheumatoid arthritis: results from ATTRACT, a multicenter, randomized, double-blind, placebo- 3 controlled trial. Arthritis Rheum 2002;46:1451–9. [5] Wolbink GJ, Vis M, Lems W, Voskuyl AE, de GE, Nurmohamed MT, et al. Development of antiinfliximab antibodies and relationship to clinical response in patients with rheumatoid arthritis. Arthritis Rheum 2006;54:711–15. 4 [6] De Vries MK, Wolbink GJ, Stapel SO, de Groot ER, Dijkmans BA, Aarden LA, et al. Inefficacy of infliximab in ankylosing spondylitis is correlated with antibody formation. Ann Rheum Dis 2007;66:133–4. 5 [7] Van der Linden S. Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8. [8] Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to 6 defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994;21:2286–91. [9] Braun J, Pham T, Sieper J, Davis J, van der LS, Dougados M, et al. International ASAS consensus statement for the use of anti-tumour necrosis factor agents in patients with 7 ankylosing spondylitis. Ann Rheum Dis 2003;62:817–24. [10] Schuurman J, Perdok GJ, Mueller GA, Benjamin DC, Yong TK, Chapman MD, et al. Mouse/ human chimeric IgG1 and IgG4 antibodies directed to the house dust mite allergen Der p 2: use in quantification of allergen specific IgG. Clin Exp Allergy 1997;27:1095–102. 8 [11] Van der Laken CJ, Voskuyl AE, Roos JC, Stigter van WM, de Groot ER, Wolbink G, et al. Imaging and serum analysis of immune complex formation of radiolabelled infliximab and anti-infliximab in responders and non-responders to therapy for rheumatoid arthritis. Ann Rheum Dis 2007;66:253–6. 9 [12] Haibel H, Brandt HC, Song IH, Brandt A, Listing J, Rudwaleit M, et al. No efficacy of subcutaneous methotrexate in active ankylosing spondylitis: a 16-week open-label trial. Ann Rheum Dis 2007;66:419–21. 10

78 79 CHAPTER 6

Elevated liver enzymes in patients with ankylosing spondylitis treated with etanercept.

J C van Denderen1 G J Blom2 I E van der Horst-Bruinsma3 B A C Dijkmans3 M T Nurmohamed1,2

Clinical Rheumatology 2012;31:1677-82.

1 Jan van Breemen Research Institute | Reade, Amsterdam. 2 VU University Medical Center, Amsterdam. 3 VU University Medical Center, Department of Rheumatology, Amsterdam. Chapter 6 Elevated liver enzymes

ABSTRACT INTRODUCTION

Objective Ankylosing spondylitis (AS) is a chronic inflammatory disease primarily affecting TNF-alpha blocking agents are very effective in patients with ankylosing spondylitis the axial skeleton and entheses. Until recently, therapeutic options were limited (AS), but several cases of liver problems have been published. We systematically to physiotherapy, nonsteroidal anti-inflammatory drugs (NSAIDs), and in case of 1 studied the frequency of this potential side effect in our AS patients treated with peripheral arthritis, sulfasalazine [1]. etanercept. Since 2002, TNF-alpha blocking agents proved to be very effective in patients Methods with spondyloarthropathy [1]. Well-known and most frequent adverse events are 2 Consecutive AS patients treated with etanercept for at least 3 months were included. infections, allergic reactions, pruritus, injection site reactions and fever [2]. Liver disease was defined as elevated liver enzymes more than 1.5 times the upper normal limit (UNL) and was categorised as probably, possibly, probably not or not In placebo-controlled trials, serious liver problems were seldom seen [3]. And thus 3 related to etanercept treatment. Patients with and without raised liver enzymes were far, a few case reports of liver disease during treatment with a TNF-alpha blocking compared for prognostic factors. agent have been published [4-7].

Results Unexpectedly, we observed in our clinical practice several AS patients with liver 4 A total of 105 patients were included. Fifteen patients had elevated liver enzymes enzyme elevations during treatment with etanercept. This prompted us to study more than once. In nine cases, the liver disease was probably (five) or possibly systematically the frequency of this potential side effect in our AS patients treated (four) related to etanercept treatment. with etanercept and to identify prognostic factors. The liver enzyme elevations were serious (>3x UNL) in six cases and resulted in 5 permanent cessation of etanercept in two cases. The nine patients with liver disease METHODS were compared with patients without elevated enzymes. No differences were found in age or use of alcohol; however, in patients with liver disease, a higher body mass From 2004 onwards, all AS patients treated with etanercept are prospectively 6 index and a trend for a higher atherogenic index were observed. Hepatic steatosis followed in the outpatients clinic of Reade (formerly Jan van Breemen Institute) in was observed in five of six patients with elevated liver enzymes. Amsterdam. All patients fulfilled the Assessment of SpondyloArthritis international Society (ASAS) criteria for treatment with a TNF-alpha blocker: AS was diagnosed Conclusion according the modified New York criteria and disease activity index (BASDAI) ≥ 4 7 Elevated serum aminotransferases, probably or possibly related to etanercept despite the use of at least two NSAIDs during at least 3 months [8]. treatment, were observed in 9% of the AS patients. An increased risk for the elevation of liver enzymes was found in patients with a higher body mass index. In these patients, laboratory tests were done at screening, baseline, week 4, week 8 We recommend regular testing of liver enzymes in patients treated with etanercept. 12 and thereafter every 3 months. The tests included aspartate aminotransferase (AST), alanine aminotransferase (ALT) and alkaline phosphatase. At screening, the following parameters are determined: gender, age, HLA-B27 status, body mass index (BMI), atherogenic index (i.e. total cholesterol/HDL-cholesterol ratio), co- 9 morbidity as diabetes mellitus, use of alcohol, all co-medication as NSAIDs, disease- modifying antirheumatic drugs (DMARDs) and isoniazide (as isoniazide is nowadays more frequnetly used in biologic treated patients). 10

For the present study, AS patients treated with etanercept 50 mg weekly for at least 3 months were included. Liver disease was defined as elevation of at least one of the liver enzymes more then 1.5 times the upper normal limit (UNL) at least at two time points. Liver enzyme elevation of ≥3 times UNL was defined as serious.

82 83 Chapter 6 Elevated liver enzymes

In patients with elevated liver enzymes, ultrasonography and relevant additional laboratory tests were done by the treating rheumatologist, such as viral hepatitis no

antigens (A, B, C, cytomegaly and Epstein-Barr), antinuclear antibody and gamma relation probably etanercept probably no glutamyl transferase. Based on this information, the authors categorised the degree possibly of relationship of the adverse events with the drug. Liver disease was categorised 1 as probably, possibly, probably not or not related to etanercept treatment based on the following definitions: probably related - the event had a strong temporal relationship to the drug or recurred on re-challenge and another aetiology was unlikely or significantly less likely; possibly related - the event had a strong temporal 2 relationship to the drug and another aetiology was equally or less likely; probably comments not related - the event had little or no temporal relationship to the drug and/or pos. rechallenge. In history ALT 110. ALT In history Same as patient 7. Same Recovery after stop after Recovery and pos. rechallenge. Strong temp. relation Strong temp. relation Strong temp. relation Strong temp. relation stop for other reason. Pos. Also Pos. rechallenge.

another aetiology was more likely; and not related - the event was due to another stop flurbiprofen. after and alcohol. Improvement after stop after Improvement Spontaneous fluctuation. Recovery after stop MTX. after Recovery Strong temp. relation and meloxicam and paroxetine. meloxicam

Infection and clindamycine. 3 influence of diclofenac / INH Improvement after stop ETN. stop ETN. after Improvement drug or an underlying or concurrent illness and was not related to etanercept (e.g. relation. Improvement Temp. Improvement after temporary after Improvement no temporal relationship or much more likely alternative aetiology; based on WHO definitions) [9]. Adjudication of the relationship was done by consensus discussion no no no no no no no no no no def. def. ETN stop stop between three of the authors (JCvD, GJB and MTN). temp. temp. temp. 4

Statistical analysis. Patients without and with raised liver enzymes, probably or possibly related to atosis n.a. n.a. n.a. n.a. n.a. n.a. n.a. scan normal

unclear. unclear. 5

the use of etanercept, were compared for baseline factors, i.e. age, sex, BMI, steatosis steatosis steatosis ste steatosis steatosis ultrasound ultrasound atherogenic index and use of alcohol. The distribution of variables was tested for steatosis CT: normality. Due to the low number of patients in the elevated ALT/AST group (n=9), a nonparametric testing method was done. Non-normally distributed variables are pos n.a. n.a. n.a. neg neg neg neg neg neg neg neg neg neg dub ANA 6 presented as medians with interquartile ranges. The groups were compared using Mann-Whitney tests. Dichotomous variables are presented as number (percentage n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. neg neg neg neg neg test of the total), and statistical significance was assessed with Fischer’s exact test. virus P values of <0.05 were considered statistically significant. All statistical analyses 7 were performed using SPSS version 17.0. 82 87 97 43 75 107 145 120 120 170 166 124 162 179 187 249 (AST 80) (AST max. ALT max. RESULTS (gGT 466) 8

A total of 105 patients were included (mean age 43 years, 74% male). Fifteen 0 1 1 3 3 2 2 1 1 1 6 1 2 1 patients (14 men) had elevated liver enzymes more than once. In nine cases, liver 10 (months) start ETN start disease was probably (five patients) or possibly (four patients) related to etanercept time after 9 treatment. In the six cases with liver disease (probably) not related to etanercept

treatment, another cause was more likely, in most cases NSAIDs use. Seven out of no no no no no no no no no no yes yes yes yes yes ALT ALT raised raised the 105 patients used isoniazide for treatment of suspected latent tuberculosis at history 10 the start of etanercept; one patient had raised liver enzymes that were only partial 43 52 58 49 36 36 26 57 50 40 37 35 63 34 27 (yr) attributable to isoniazide. One of the 15 patients had diabetes mellitus. age f f m m m m m m m m m m m m m Clinical details of the 15 ankylosing spondylitis patients with elevated liver enzymes (ALT or AST >1.5 UNL) during treatment with etanercept. >1.5 AST or ankylosingClinical details of the 15 spondylitis patients with elevated liver enzymes (ALT In table 1, details of the patients with elevated liver enzymes are shown. In all sex but one cases, the increase in ALT was higher than in AST. Levels of alkaline 1 2 3 4 5 6 7 8 9 11 nr 10 14 12 13 15 Table 1. 1. Table maximal level of ALT: etanercept. Max. start ETN: time of event in months after start Time after ULN in past. >1.5 AST or ALT history: ALT Raised Dub: dubious. Virus test: antibody tests for viral hepatitis. N.a.: not applicable. Neg: negative. transferase (U/l). glutamyl gamma (U/l). gGT: ALT Stop ETN: cessation of treatment with etanercept computer tomography. Steatosis: hepatic steatosis. CT: ANA: antinuclear antibody. positive. Pos: Temp. positive re-challenge. Pos. rechall.: MTX: methotrexate. INH: isoniazide. Def: definitively. temporary. Temp: because of increased liver enzymes. relation: temporal relation.

84 85 Chapter 6 Elevated liver enzymes

phosphatase were (nearly) normal and gamma glutamyl transferase was elevated Table 2. Median baseline values of patients with liver disease probably or possibly in one patient. In the 2 years before start of etanercept, at least one to three results related to the use of etanercept compared to patients using etanercept without of liver enzyme tests were known of most patients. In the five cases with “liver raised liver enzymes. disease probably related to etanercept,” the liver enzymes were normal before the normal ALT/AST elevated ALT/AST p-value* start of etanercept. In three of the four cases with “liver disease possibly related (n=90) (n=9) 1 to etanercept”, liver enzymes were abnormal at the start of etanercept and two had Age (years), med (IQR) 43 (37-51) 38 (36-51) 0.679 elevated ALT >1.5 ULN in the past. However, in these cases, the enzymes clearly Male, no (%) 64 (72) 7 (78) 1.000 increased after start of the medication and etanercept seemed to aggravate a pre- BMI, med (IQR) 25 (23-28) 31 (29-34) 0.000 Atherogenic index, med (IQR) 3.6 (2.9-4.5) 4.3 (3.7-4.6) 0.151 existing liver problem. DMARDs (3x sulfasalazin, 1x methotrexate) were stopped Alcohol (units/wk), med (IQR) 1.0 (0.0-7.0) 2.0 (0.0-9.0) 0.885 2 because of the elevated enzymes, resulting in recovery in only one case. Thirteen of the 15 patients used a NSAID, which was stopped to assess the possible effect *Mann-Whitney test; Male: Fischer’s exact test. Med (IQR): median (inter quartile range). BMI: body mass index (kg/m2). Atherogenic index: total cholesterol/HDL-cholesterol ratio. on the liver enzymes. In some cases, the liver disease appeared to be related to the combination of medication, as liver enzymes were raised during the use of the 3 combination of etanercept and other medication and were normal during the use of Figure 1 depicts the levels of ALT and AST, in relation to the medication, in a the individual drugs. Increase of liver enzymes was observed 1-3 months after the representative HLA-B27 positive male patient aged 34 years with elevated liver start of etanercept, but in one case after 6 months. enzymes probably related to etanercept (patient 14 in table 1). He used naproxen 4 for several years. After starting etanercept 25 mg twice weekly, ALT and, to a lesser Hepatic steatosis, diagnosed by ultrasonography and in one case by computer extent, AST raised in 3 months from near normal levels to 157 U/L (normal <45) tomography (CT), was observed in most patients (seven out of eight) with elevated and 80 U/L (normal <35), respectively. Alkaline phosphatase and gamma glutamyl liver enzymes. Ultrasonography was only performed when the treating physician 5 transferase were normal. IgM antibodies to Epstein-Barr, cytomegalovirus and considered this as necessary and therefore was not done in all patients with liver hepatitis A were negative as well as HBsAg, anti-HBs, hepatitis C antibodies and disease and in none of the patients without liver abnormality. No liver biopsy was the antinuclear antibody (ANA) test. Baseline total cholesterol was 3.7 mmol/l, the done. atherogenic index was 4.6 and he was overweight (BMI 39 kg/m²). CT scan revealed 6 hepatic steatosis. Due to the presence of obesity, a liver biopsy was not performed. Serious liver enzyme elevations (≥3 times UNL) were observed in six cases and There was no alcohol abuse and no co-morbidity such as diabetes mellitus. resulted in definitive cessation of etanercept in two cases with decreasing enzymes Stopping naproxen did not change the level of the liver enzymes. After stopping during the follow-up period. In the other cases, the liver enzymes decreased after 7 etanercept, ALT dropped to 88 U/l in 2 months. Another 2 months later, etanercept temporary cessation or after stopping other medication or the enzymes fluctuated in was restarted and ALT rose again, from 75 U/l to 179 U/l in one month. After again ranges that were considered acceptable by the treating rheumatologist. stopping the etanercept, ALT dropped to 122 U/l in a few weeks and after 3 months ALT was 66 U/l. The figure also shows that liver enzymes raised again after the start 8 The nine patients with liver disease probably or possibly related to etanercept were of adalimumab, another TNF-alpha blocking agent. compared with the patients without elevated enzymes (table 2). No differences were found with respect to age, sex or alcohol use. However, in patients with liver disease, a higher BMI (p=0.000) and a trend for a higher atherogenic index (i.e. total 9 cholesterol/HDL-cholesterol ratio) were observed. (p=0.15) 10

86 87 Chapter 6 Elevated liver enzymes

The second study was a placebo-controlled trial of golimumab in AS. Increased 250 etanercept adalimumab (>UNL) ALT and AST levels were seen in 2.6 and 1.3% of the patients in the placebo NSAID group and 6.0 and 5.3% in the golimumab-treated patients. Marked abnormal post- baseline ALT or AST values were observed in one placebo-treated patient and in 200 eight golimumab-treated patients. In three patients (all in the 100 mg group), the 1 study drug was discontinued for this reason.

150 In summary, in these placebo-controlled trials, liver enzyme increases of >3x UNL ALAT are seen occasionally and more frequent in the patients using the TNF blocker. 2 ASAT 100 Also, the mean increase of liver enzymes is higher compared to the placebo-treated patients. 3 50 In our study, the frequency of elevation of liver enzymes in patients treated with etanercept was investigated. However, these elevations have been seen in all TNF blocking agents, as was described in the above-mentioned placebo-controlled trials 0 and in several published case histories [4, 5]. It is unknown if the frequency of this 4 -28 -24 -20 -16 -12 -8 -4 0 4 8 12 16 20 adverse event differs between the several TNF-blocking agents in AS. In a study in

patients with rheumatoid arthritis, liver enzyme elevations were most consistently Figure 1. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels Figure(U/l) and 1. useSerum medication alanine aminotransferasein a male with ankylosing (ALT) and spondylitis. aspartate X-axis aminotransferase time in months. (AST) levels associated with infliximab, less commonly observed with adalimumab and not (U/l) and use medication in a male with ankylosing spondylitis. X-axis time in months. observed with etanercept compared with comparator DMARDs [15]. Unlike our case 5 shown in the figure, it has been reported that a successful switch to another TNF DISCUSSION blocker is possible without relapse [6, 16]. Perhaps the frequency of elevation of liver enzymes during treatment with a TNF blocker is higher in AS than in other rheumatic 6 Increased liver enzymes were found in 15 of 105 (14%) AS patients who were treated diseases [17]. Obviously, this cannot be explained by the use of hepatotoxic co- with etanercept for at least 3 months. In nine of these cases (60%), a relation of medication like methotrexate, which is used more often in rheumatoid arthritis and liver enzyme abnormality with the use of etanercept was classified as probable or seldom in AS. A role of other co- as NSAIDs and prednisone is unlikely, possible. but cannot be completely excluded. 7

As raised liver enzymes are observed in studies of the normal population, When elevation of liver enzymes is observed during treatment with medication, comparison with a placebo group would be preferable [10]. However, we compared other causes like alcohol and other toxic agents, viral infections, co-medication and 8 the data of the patients with the levels of liver enzymes in the 2 years before co-morbidity have to be excluded. When another aetiology is unlikely and the event the start of etanercept. In several placebo-controlled trials with TNF blockers, recurs on re-challenge, a relation with the drug is probable. Several pathogenetic liver enzyme abnormalities were seen sporadically, but here, only the more mechanisms could play a role. severe abnormalities were mentioned [3, 11, 12]. Two studies however, which are 9 summarized below, reported less severe abnormalities as well [13, 14]. One possibility is reactivation of viral hepatitis which was described in patients The first study was a double-blind trial comparing adalimumab with placebo receiving etanercept [18]. However, in other case series of patients with viral for 24 weeks in patients with AS. Mean increases in ALT, AST and total bilirubin hepatitis, etanercept appeared to be safe [19-23]. Another option is that toxic 10 concentrations were statistically significant different between the two groups. hepatitis occurs due to etanercept like the reported case of liver biopsy-confirmed Six patients taking adalimumab and one taking placebo had a post-baseline ALT toxic hepatitis attributable to infliximab [6]. Another publication described a liver concentration ≥3x UNL. The ALT concentration returned to normal during continued biopsy which revealed a granulomatous hepatitis associated with etanercept [5]. adalimumab treatment in four of the six patients. An AST concentration ≥3x UNL This case is interesting because more cases of sarcoid-like granulomatosis were was less frequently observed. reported in patients treated with TNF blockers [24]. Theoretically, this is related to the

88 89 Chapter 6 Elevated liver enzymes

type of TNF blocker and seen more often in the soluble receptor blocker compared In conclusion, elevated serum aminotransferases, probably or possibly related to to the monoclonal antibodies. This might be caused by differences in the action on etanercept treatment, were observed in 9% of the AS patients. An increased risk for T cells and differences in the ability to induce apoptosis [25]. This different effect elevation of liver enzymes was found in patients with a high body mass index and on granuloma formation is also reflected in the higher risk of tuberculosis with high atherogenic index. monoclonal antibody therapy [25]. 1 REFERENCES TNF blockers can also induce autoimmune disease by triggering development of auto-antibodies and following infliximab therapy autoimmune hepatitis has [1] Zochling J, Van der Heyde D, Burgos-Vargas R, Collantes E, Davis Jr JC, Dijkmans B, et been diagnosed based on liver biopsy and positive ANA and double-stranded al. ASAS/EULAR recommendations for the management of ankylosing spondylitis. Ann 2 Rheum Dis 2006;65:442-52. DNA antibody test results [4]. These authors suggested the possibility that this http://www.ema.europa.eu/humandocs/PDFs/EPAR/Enbrel/emea-combined-h262nl.pdf. adverse event could be more frequent in AS than in rheumatoid arthritis, as in [2] Accessed September 2009. AS immunosuppressive co-medication is used much less frequent. In case of [3] Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosing spondylitis by inhibition of autoimmune hepatitis, therapy with steroids can be successful. As in our study, 3 tumor necrosis factor alpha. N Engl J Med 2002;346:1349-56. in the published cases and in the above-mentioned placebo-controlled trials, [4] Ozorio G, McGarity B, Bak H, Jordan AS, Lau H, Marshall C. Autoimmune hepatitis. laboratory tests in TNF blockers suggest hepatocellular injury. However, in rare cases following infliximab therapy for ankylosing spondylitis. Med J Aust 2007;187:524-6. cholestatic liver disease is described [5, 26]. [5] Farah M, Al Rashidi A, Owen DA, Yoshida EM, Reid GD. Granulomatous hepatitis 4 associated with etanercept therapy. J Rheumatol 2008;35:349-51. In our study, an increase of liver enzymes was associated with a significantly higher [6] Carlsen KM, Riis L, Madsen OR. Toxic hepatitis induced by infliximab in a patient with body mass index and a higher atherogenic index. Serum ALT activity is found to be rheumatoid arthritis with no relapse after switching to etanercept. Clin Rheumatol independently related to body mass index and to laboratory indicators of abnormal 2009;28:1001-3. 5 lipid or carbohydrate metabolism [27]. Non-alcoholic fatty liver disease (NAFLD) [7] Tobon GJ, Cañas C, Jaller JJ, Restrepo JC, Anaya JM. Serious liver disease induced by is the most common cause of elevated liver enzymes in the developed countries infliximab. Clin Rheumatol 2007;26:578-81. [28]. Most NAFLD patients will develop impaired glucose tolerance and progression [8] Braun J, Pham T, Sieper J, Davis J, Van der Linden S, Dougados M, et al. International 6 of liver fibrosis is associated with more pronounced insulin resistance and more ASAS consensus statement for the use of anti-tumour necrosis factor agents in patients with ankylosing spondylitis. Ann Rheum Dis 2003;62:817-24. significant weight gain. Several cytokines play a role in these processes. An increased expression of inflammatory enzymes like TNF-alpha is found in alcoholic [9] http://www.who-umc.org/graphics/4409.pdf. Accessed September 2009. and non-alcoholic fatty liver disease [29]. TNF blockers may also have a favourable [10] Ioannou GN, Weiss NS, Boyko EJ, Mozaffarian D, Lee SP. Elevated serum alanine 7 aminotransferase activity and calculated risk of coronary heart disease in the United effect on the lipid profile in patients with rheumatic diseases [30]. Therefore, States. 2006;43:1145-51. blocking TNF-alpha could have a favourable effect in fatty liver disease. But our [11] Calin A, Dijkmans BAC, Emery P, Hakala M, Kalden J, Leirisalo-Repo M, et al. Outcomes of study suggests that etanercept can aggravate a pre-existing liver problem, possibly a multicentre randomised clinical trial of etanercept to treat ankylosing spondylitis. Ann 8 because hepatocytes in livers with fatty changes are more vulnerable to the effects Rheum Dis 2004;63:1594-600. of TNF blockers. [12] Van der Heijde D, Da Silva JC, Dougados M, Geher P, Van der Horst-Bruinsma IE, Juanola X, et al. Etanercept 50 mg once weekly is as effective as 25 mg twice weekly in patients In most patients, a small increase of liver enzymes has no clinical relevance. with ankylosing spondylitis. Ann Rheum Dis 2006;65:1572-7. 9 However, in several of our patients, we had to stop co-medication, had to start [13] Van der Heijde D, Kivitz A, Schiff MH, Sieper J, Dijkmans BA, Braun J, et al. Efficacy and another TNF blocker and in some cases anti-TNF therapy had to be stopped safety of adalimumab in patients with ankylosing spondylitis: results of a multicenter, permanently. In the literature, several cases of severe hepatic toxicity have been randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2006;54:2136-46. 10 reported including jaundice, hepatic failure, liver transplantation and death [4]. [14] Inman RD, Davis JC Jr, Van der Heijde D, Diekman L, Sieper J, Kim SI, et al. Efficacy and safety of golimumab in patients with ankylosing spondylitis: results of a randomized, Therefore, we recommend regular testing of liver enzymes, especially ALT and AST, double-blind, placebo-controlled, phase III trial. Arthritis Rheum 2008;58:3402-12. in all patients treated with TNF blockers, particularly in those patients that seem [15] Sokolove J, Strand V, Greenberg JD, Curtis JR, Kavanaugh A, Kremer JM, et al. Risk to be more at risk, i.e. patients with a high body mass index and high atherogenic of elevated liver enzymes associated with TNF inhibitor utilisation in patients with index. rheumatoid arthritis. Ann Rheum Dis 2010;69:1612-7.

90 91 Chapter 6 Elevated liver enzymes

[16] Massarotti M, Marasini B. Successful treatment with etanercept of a patient with psoriatic arthritis after adalimumab-related hepatotoxicity. Int J Immunopathol Pharmacol 2009;22:547-9. [17] http://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?id=8421; table 11. Accessed September 2009. [18] Cansu DU, Kalifoglu T, Korkmaz C. Short-term course of chronic hepatitis B and C under 1 treatment with etanercept associated with different disease modifying antirheumatic drugs without antiviral prophylaxis. J Rheumatol 2008;35:421-4. [19] Peterson JR, Hsu FC, Simkin PA, Wener MH. Effect of tumour necrosis factor alpha antagonists on serum transaminases and viraemia in patients with rheumatoid arthritis 2 and chronic hepatitis C infection. Ann Rheum Dis 2003;62:1078-82. [20] Oniankitan O, Duvoux C, Challine D, Mallat A, Chevalier X, Pawlotsky JM, et al. Infliximab therapy for rheumatic diseases in patients with chronic hepatitis B or C. J Rheumatol 2004;31:107-9. 3 [21] Parke FA, Reveille JD. Anti-tumor necrosis factor agents for rheumatoid arthritis in the setting of chronic hepatitis C infection. Arthritis Rheum 2004;51:800-4. [22] Roux CH, Brocq O, Breuil V, Albert C, Euller-Ziegler L. Safety of anti-TNF-alpha therapy in 4 rheumatoid arthritis and spondylarthropathies with concurrent B or C chronic hepatitis. Rheumatology 2006;45:1294-7. [23] Marotte H, Fontanges E, Bailly F, Zoulim F, Trepo C, Miossec P. Etanercept treatment for three months is safe in patients with rheumatological manifestations associated with 5 hepatitis C virus. Rheumatology 2007;46:97-9. [24] Daïen CI, Monnier A, Claudepierre P, Constantin A, Eschard JP, Houvenagel E, et al. Sarcoid-like granulomatosis in patients treated with tumor necrosis factor blockers: 10 cases. Rheumatology 2009;48:883-6. 6 [25] Tubach F, Salmon D, Ravaud P, Allanore Y, Goupille P, Bréban M, et al. Risk of tuberculosis is higher with anti-tumor necrosis factor monoclonal antibody therapy than with soluble tumor necrosis factor receptor therapy: The three-year prospective French Research Axed on Tolerance of Biotherapies registry. Arthritis Rheum 2009;60:1884-94. 7 [26] Menghini VV, Arora AS. Infliximab-associated reversible cholestatic liver disease.Mayo Clin Proc 2001;76:84-6. [27] Prati D, Taioli E, Zanella A, Della Torre E, Butelli S, Del Vecchio E, et al. Updated definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med 2002;137:1- 8 10. [28] Ekstedt M, Franzén LE, Mathiesen UL, Thorelius L, Holmqvist M, Bodemar G, et al. Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006;44:865-73. 9 [29] Chu CJ, Lu RH, Wang SS, Chang FY, Wu SL, Lu CL, et al. Risk factors associated with non- alcoholic fatty liver disease in Chinese patients and the role of tumor necrosis factor-alpha. Hepatogastroenterology 2007;54:2099-102. 10 [30] Jamnitski A, Visman IM, Peters MJ, Dijkmans BA, Voskuyl AE, Nurmohamed MT. Beneficial effect of 1-year etanercept treatment on the lipid profile in responding patients with rheumatoid arthritis: the ETRA study. Ann Rheum Dis 2010;69:1929-33.

92 93 CHAPTER 7

What do we miss? ASAS non-responders on anti-TNF therapy show improvement in performance-based physical function.

S F E van Weely1 J C van Denderen1 M P M Steultjens2 M T Nurmohamed1 B A C Dijkmans1,3 J Dekker1,4,5 I E van der Horst-Bruinsma3

Rheumatology (Oxford) 2013;52:1884-9.

1 Jan van Breemen Research Institute | Reade, Amsterdam. 2 Institute for Applied Health Research and School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, UK. 3 Department of Rheumatology, VU University Medical Center, Amsterdam. 4 Department of Rehabilitation Medicine and EMGO Institute, VU University Medical Center, Amsterdam. 5 Department of Psychiatry, VU University Medical Center, Amsterdam. Chapter 7 Performance-based physical function

ABSTRACT INTRODUCTION

Objective AS is characterized by limitations in physical functioning due to pain, stiffness A prospective study was conducted in order to establish whether AS patients, and fusion of the spine. Anti-TNF therapy has been shown to improve physical who are defined as non-responders after 3 months of anti-TNF therapy, show functioning [1]. For evaluation of the disease course and the effectiveness of anti- 1 improvement on performance-based tests of physical functioning. TNF therapy, physical functioning is an important outcome measure.

Methods Physical functioning in AS is most commonly assessed with the BASFI [2, 3], At baseline and 3 months after the start of anti-TNF therapy, AS patients completed a self-reported, disease-specific, valid and reliable outcome measure [4–6]. In 2 seven performance-based tests of physical functioning, questionnaires on self- the absence of a true gold standard, the BASFI is considered the best option to reported physical functioning (BASFI) and disease activity (BASDAI), and a pain assess physical functioning. However, it is a self-reported outcome measure and and a global patient assessment. The concordance between ≥20% intra-individual therefore is susceptible to subjective interpretation (under- or overestimation) due 3 improvement on the performance-based test of physical functioning and (i) to confounding effects of perceived physical functioning, personality traits, pain, response to anti-TNF therapy [Assessment of SpondyloArthritis international language or depression [7–10]. Society 20% (ASAS20) response] and (ii) ≥20% intra-individual improvement on self- reported physical functioning (BASFI) was assessed. Performance-based tests are a more objective outcome measure to evaluate 4 physical functioning. Performance-based tests of physical functioning based on Results the BASFI have been developed and have shown adequate to excellent reliability One hundred AS patients were included, of which 82 patients completed all [11]. The association between the performance-based tests and the BASFI is only tests at both time points. After 3 months of anti-TNF therapy, 27 (32.9%) patients moderate [11, 12]. Furthermore, a previous study showed that alongside actual 5 were categorized as non-responders according to the ASAS20 response criteria. performance, AS patients seem to incorporate exertion and pain in their assessment Improvement in performance-based physical functioning was seen in 13 of the 27 of perceived physical functioning on the BASFI [12]. This suggests that performance non-responders (48.1%) (i.e. n = 13/82 = 15.9% of the total group). Furthermore, and self-reported measures do not measure the same aspects of physical 6 30 (36.6%) patients showed no improvement on self-reported physical functioning functioning. Consequently performance-based tests could provide an objective (BASFI). However, 17 of the 30 (56.7%) patients did improve on the performance- outcome measurement for the evaluation of physical functioning and give additional based tests of physical functioning (i.e. n = 17/82 = 20.7% of the total group). information on changes in physical functioning in addition to the BASFI. This would provide an argument for the use of performance-based tests alongside the BASFI in 7 Conclusion the evaluation of treatment modalities like anti-TNF therapy. After 3 months of anti-TNF therapy, performance-based tests of physical functioning showed improvement in 48.1% of the ASAS20 non-responders. With these This prospective study therefore aimed to establish whether AS patients showed 8 performance-based tests, new information on outcome after anti-TNF therapy can improvement on performance-based tests of physical function after 3 months of be generated. Using performance-based tests alongside the BASFI could have anti-TNF therapy (etanercept or adalimumab). We investigated whether patients, additional value in the evaluation of outcomes for patients receiving anti-TNF defined as non-responders according to the Assessment of SpondyloArthritis therapy. International Society 20% response (ASAS20) [13, 14], showed improvement 9 in performance-based physical functioning. Furthermore, we investigated the differences between improvement in performance-based and self-reported (BASFI) physical functioning after 3 months of anti-TNF therapy. 10

96 97 Chapter 7 Performance-based physical function

PATIENTS AND METHODS (vii) Physically demanding activities: two pylons were placed 10 m apart. Patients were provided with a heart rate monitoring device and were instructed to Patients were recruited from a large outpatient centre for rheumatology and perform the shuttle walk test. The test was stopped if the patient’s heart rehabilitation (Reade) in Amsterdam. Enrolment took place from May 2006 to June frequency (HF) exceeded 80% of the HF maximum, if the patient could not keep 2010. The following inclusion criteria were applied: diagnosis of AS according to the up with the pace as instructed by the assessor or if the patient wanted to stop. 1 modified New York criteria [15], ≥18 years of age, eligible for treatment with anti-TNF and sufficient command of the Dutch language. Patients were excluded if they had Statistical analyses pulmonary, cardiovascular or neurological comorbidity affecting the patient’s ability Descriptive statistics were computed by calculating mean and s.d. for all continuous to perform daily activities. The study was approved by the medical ethics committee data and percentages for categorical data. Paired samples t-tests were used to 2 of Reade. All patients gave written informed consent according to the Declaration of assess improvement on performance-based tests and self-reported physical Helsinki. functioning (BASFI), disease activity (BASDAI) and spinal and hip mobility (BASMI). 3 Measures A score for performance-based physical functioning was computed by calculating At baseline and after 3 months of anti-TNF treatment (etanercept or adalimumab), the mean of the seven performance-based tests [12]. Although the unit of patients completed seven performance-based tests of physical functioning, measurement for all tests was the same (i.e. seconds), a standardization procedure questionnaires on self-reported physical functioning (BASFI) and disease activity was necessary because the distributions between the tests varied (i.e. different 4 (BASDAI) [16], and a pain and patient global assessment. At both time points, spinal mean and s.d.). Therefore raw performance scores were transformed into z-scores. and hip mobility was also assessed, using the BASMI [17, 18]. In this way, all tests contributed evenly to the mean performance score.

Before each test the patient was uniformly instructed as to how to execute the test. The ASAS20 response [13, 14] was used to categorize patients as responders or 5 The tests were carried out in the following order. The outcome of the performance non-responders to anti-TNF therapy. Analogous to the ASAS criteria, an intra- tests was the time needed to complete the task, measured in seconds. A detailed individual improvement of ≥20% was used to classify patients as improvers or non- description of the performance-based test was given in an earlier publication [11]. improvers on performance-based tests of physical functioning. On self-reported 6 The performance tests were based on items of the BASFI and consisted of seven physical functioning, patients were defined as improved if they showed an intra- items representing one domain: physical functioning [12]. individual improvement of ≥20% and ≥1 unit on the BASFI. This is the same extent of (i) Climbing stairs: patients faced a flight of 12 steps and were instructed to climb improvement as described for the BASFI in the ASAS20 improvement criteria [13, 14]. the stairs without using the handrail or a walking aid. 7 (ii) Bending: patients were instructed to bend forward from the waist and pick up Subsequently, cross-tabulations were produced to establish the concordance six pens from the floor without an aid and place them on a shelf one by one. between improvers on performance-based physical functioning (number and (iii) Reaching: patients faced two shelves placed below each other. Six pens were percentage) in (i) (non-) responders on the ASAS20 criteria and (ii) (non-)improvers 8 placed on the lower shelf. Patients were instructed to reach up and place the on self-reported physical functioning (BASFI). All analyses were performed using pens on the highest shelf without help or aids. SPSS for Windows 18.0 (SPSS Inc., Chicago, IL, USA). (iv) Putting on socks: patients stood barefooted with a pair of socks in one hand and were instructed to put on the socks without help or aids. RESULTS 9 (v) Reclining and declining from a chair: patients were instructed to stand up and sit down three times in a row from the chair without using their hands or any other Baseline characteristics assistance. The study population consisted of 100 patients with a confirmed diagnosis of AS, 10 (vi) Getting up from the floor: patients began the test lying supine on the mat and fulfilling the modified New York criteria. Data for 18 patients were incomplete after were instructed to get up without help. follow-up; 7 patients (7%) did not start treatment, 4 stopped treatment within 3 months (4%) and 7 (7%) were not reassessed due to other reasons. No patients were excluded for having pulmonary, cardiovascular or neurological comorbidity affecting the patient’s ability to perform daily activities. Accordingly, 82 patients (67.5% men,

98 99 Chapter 7 Performance-based physical function

n = 56) with a mean age ( ± s.d.) of 43.9 (±11.3) years were reassessed and included Improvement in performance-based tests vs ASAS20 response after anti-TNF in the analyses. Medication at baseline was used by 89% of the study population therapy and consisted mainly of NSAIDs. Table 1 displays the characteristics of the study Table 2 shows the cross-tabulation between (non-)improvers on the performance- population. based tests and (non-)responders on the ASAS20 response. Improvement in performance-based physical functioning was seen in 70.9% (n = 39/55) of the 1 According to the ASAS20 criteria, 27 (32.9%) patients were categorized as non- responders (i.e. n = 39/82 = 47.6% of the total group). However, 48.1% (n = 13/27) of responders. After 3 months of anti-TNF therapy, patients experienced a significant the ASAS20 non-responders showed a ≥20% improvement in performance-based decline in self-reported limitations in physical functioning and disease activity, physical functioning (i.e. n = 13/82 = 15.9% of the total group). as shown by lower BASFI and BASDAI scores (P < 0.0001). Furthermore, an 2 improvement in spinal mobility (BASMI) (P < 0.01) and performance-based physical functioning were observed (P < 0.0001). Table 2. (Non-)improvers (performance-based functioning) vs (non-)responders (ASAS20) and (non-)improvers (self-reported functioning) after 3 months anti-TNF therapy (n = 82)a. 3 Table 1. Characteristics of 82 AS patients. Improvement in performance-based After 3 months of anti-TNF Before anti-TNF therapy physical functioningb, n (%) therapy No Ye s Total 4 Men, % (n) 67.5 (56) Responder to anti-TNF therapyc (ASAS20 No 14 (17.1) 13 (15.9) 27 (32.9) Agea, years 43.9 (11.3) response) Yes 16 (19.5) 39 (47.6) 55 (67.1) Symptom durationa, years 21.3 (11.1) No 13 (15.9) 17 (20.7) 30 (36.6) Disease durationa, years 13.4 (9.4) Improvement in self-reported physical 5 functioningd (BASFI) Yes 17 (20.7) 35 (42.7) 52 (63.4) Medication, % (n) None 9.6 (8) Total 30 (36.6) 52 (63.4) 82 (100) NSAIDs 77.1 (64) aPercentages of the total group are given. bImprovement defined as ≥20% intra-individual 6 DMARDsb 12.0 (10) change on performance-based tests. cResponder defined according to ASAS20 response crite- ria: improvement of ≥20% and ≥1 unit in at least three of four self-reported domains, and no HLA-B27+, % (n) 79.5 (66) worsening of ≥20% and ≥1 unit in the remaining domain on an NRS of 0–10. dImprovement ESRa, mm/h 22.8 (19.0) defined as ≥20% and ≥1 unit intra-individual change on the BASFI questionnaire. Extra-spinal symptoms 7 Arthritis, % (n) 39.8 (33) Inflammatory bowel disease, % n( ) 6.0 (5) Improvement in performance-based vs self-reported physical functioning after anti- Psoriasis, % (n) 7.2 (6) TNF therapy 8 Uveitis, % (n) 30.1 (25) The cross-tabulation between (non-)improvers on the performance-based tests and ASAS20 response, % (n) (non-) improvers on self-reported physical functioning (BASFI) is also shown in Table Responders 67 (55) 2. Thirty patients (36.6%) did not improve in physical functioning according to the Non-responders 33 (27) BASFI after 3 months of anti-TNF therapy. However, 17 of the 30 patients (56.7%) 9 a BASFI , score 0–10 5.4 (2.1) 3.3 (2.4) P < 0.0001 showed an improved performance-based physical functioning (i.e. n = 17/82 = 20.7% BASDAIa, score 0–10 6.0 (1.7) 3.1 (2.1) P < 0.0001 of the total group). BASMIa, score 0–10 4.1 (1.9) 3.7 (1.9) P < 0.01 10 Performance-based testsa,c 1.03 (0.84) 0.72 (0.60) P < 0.0001 Of the total group (n = 82), 57 patients were treated with etanercept and 25 patients aValues are presented as mean (s.d.), unless indicated otherwise. bDMARDs: SSZ, MTX. cRep- were treated with adalimumab. Analyses were repeated and performed separately orted score is mean z-score. for etanercept and adalimumab. Analyses yielded similar results to the total group.

100 101 Chapter 7 Performance-based physical function

DISCUSSION The question arises as to what extent measurement error can explain changes in performance-based tests. In an earlier publication we showed that the reliability This prospective study focused on the improvement in performance-based physical of the performance-based measures is adequate to excellent (ICCs 0.73–0.96) and functioning after 3 months of anti-TNF therapy. Our results showed that 48.1% of the no systematic differences in the measurements were observed [11]. Therefore it is AS patients, who were defined as non-responders on the ASAS20 criteria, displayed unlikely that an intra-individual improvement of 20% or more is only attributable to 1 at least 20% improvement in performance-based physical functioning (i.e. 15.9% measurement error. of the total group). Furthermore, 56.7% of the patients who did not improve in self- reported physical functioning (BASFI) did improve on the performance-based tests We propose that evaluating physical functioning is not a matter of choosing one (i.e. 20.7% of the total group). best measurement method. Rather, we would prefer to identify the best combination 2 of measurements that reflects the patients’ perspective (i.e. questionnaires) as Until now, outcome after anti-TNF therapy had only been measured with self- well as true ability in the physical function domain (i.e. performance-based tests). reported outcomes like the BASFI, BASDAI 50% response [19] or a combination of Using a combination of the BASFI and performance-based tests may provide a 3 self-reported outcomes such as the ASAS20 response (which contains the BASFI). better reflection of changes in the physical function domain in AS patients. Patients Our results suggest that using both self-reported and performance-based outcome might be denied continuation of anti-TNF treatment despite their improvement, measures could have additional value in evaluating outcome after interventions in or they may not be given access to physiotherapy despite limitations in physical AS patients. functioning. 4

We showed that 48.1% of the non-responders to anti-TNF therapy (ASAS20) In conclusion, after 3 months of anti-TNF therapy, improvement in performance- improved in performance-based physical functioning. This suggests that patients based physical functioning was seen in 48.1% of the ASAS20 non-responders. improve due to the effects of anti-TNF therapy objectively (e.g. they can get up from Performance-based tests provide the opportunity to generate new information in 5 the floor more quickly), however, they do not feel better in a subjective way (i.e. the evaluation of outcome after anti-TNF therapy. In the future, using performance- they still experience pain, stiffness and/or fatigue). We also would like to point out based tests alongside the BASFI could provide additional value in evaluating that 29.1% (n = 16/55) of the patients who showed a response according to ASAS outcome for AS patients receiving anti-TNF therapy. 6 criteria did not improve on the performance-based tests. This suggests that patients experience a decline in pain, stiffness or fatigue and/or improvement in self-reported physical functioning after anti-TNF therapy, while the actual level of physical functioning may not have changed at all. Both observations are in line with previous 7 findings that self-reported levels of functioning are strongly related to pain and exertion and less to the time required to complete a task [10, 12]. Also, the contrast between improvement in self-reported (BASFI) and performance-based physical 8 functioning underlines that perceived and actual physical function are two related but distinct entities.

In this study, analogous to the ASAS criteria, a ≥20% intra-individual improvement 9 in performance-based physical functioning was used to categorize patients as improvers or non-improvers. The advantage of this approach is that relative, intra- individual changes may provide a better reflection of clinically meaningful changes 10 for individual patients than absolute changes. In rheumatology, this is a commonly used approach for defining improvement [20]. Also, in the ASAS20 or BASDAI50 criterion [13, 14, 19], an intra-individual change in terms of percentage (i.e. 20% or 50%) for defining improvement after anti-TNF therapy is used.

102 103 Chapter 7 Performance-based physical function

REFERENCES

[1] Escalas C, Trijau S, Dougados M. Evaluation of the treatment effect of NSAIDs/TNF [17] Jenkinson TR, Mallorie PA, Whitelock HC, et al. Defining spinal mobility in ankylosing blockers according to different domains in ankylosing spondylitis: results of a meta- spondylitis (AS). The Bath AS Metrology Index. J Rheumatol 1994;21:1694-8. analysis. Rheumatology 2010;49:1317-25. [18] Jones SD, Porter J, Garrett SL, et al. A new scoring system for the Bath Ankylosing [2] Calin A, Garrett S, Whitelock H, et al. A new approach to defining functional ability in Spondylitis Metrology Index (BASMI). J Rheumatol 1995;22:1609. 1 ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional [19] Braun J, Pham T, Sieper J, et al. International ASAS consensus statement for the use of Index. J Rheumatol 1994;21:2281-5. anti-tumour necrosis factor agents in patients with ankylosing spondylitis. Ann Rheum Dis [3] Van Tubergen A, Debats I, Ryser L, et al. Use of a numerical rating scale as an answer 2003;62:817-24. modality in ankylosing spondylitis-specific questionnaires.Arthritis Rheum 2002;47:242-8. [20] Tubach T, Ravaud P, Martin-Mola E, et al. Minimum clinically important improvement and 2 [4] Auleley GR, Benbouazza K, Spoorenberg A, et al. Evaluation of the smallest detectable patient acceptable symptom state in pain and function in rheumatoid arthritis, ankylosing difference in outcome or process variables in ankylosing spondylitis. Arthritis Rheum spondylitis, chronic back pain, hand osteoarthritis, and hip and knee osteoarthritis: results 2002;47:582-7. from a prospective multinational study. Arthritis Care Res 2012;64:1699-707. 3 [5] Calin A, Nakache JP, Gueguen A, et al. Outcome variables in ankylosing spondylitis: evaluation of their relevance and discriminant capacity. J Rheumatol 1999;26:975-9. [6] Ruof J, Stucki G. Comparison of the Dougados Functional Index and the Bath Ankylosing Spondylitis Functional Index. A literature review. J Rheumatol 1999;26:955-60. 4 [7] Guralnik JM, Branch LG, Cummings SR, et al. Physical performance measures in aging research. J Gerontol 1989;44:M141-6. [8] Hoeymans N, Feskens EJ, van den Bos GA, et al. Measuring functional status: cross- sectional and longitudinal associations between performance and self-report (Zutphen 5 Elderly Study 1990–1993). J Clin Epidemiol 1996;49:1103-10. [9] Kivinen P, Sulkava R, Halonen P, et al. Self-reported and performance-based functional status and associated factors among elderly men: the Finnish cohorts of the Seven Countries Study. J Clin Epidemiol 1998;51:1243-52. 6 [10] Terwee CB, van der Slikke RMA, van Lummel RC, et al. Self-reported physical function was more influenced by pain than performance-based physical function in knee-osteoarthritis patients. J Clin Epidemiol 2006;59:724-31. [11] Van Weely SFE, van Denderen JC, van der Horst-Bruinsma IE, et al. Reliability of 7 performance measures of physical function based on the BASFI, in ankylosing spondylitis. Rheumatology 2009;48:1254-60. [12] Van Weely SFE, van Denderen JC, Steultjens MPM, et al. Moving instead of asking? Performance-based tests and BASFI questionnaire measure different aspects of physical 8 function in ankylosing spondylitis. Arthritis Res Ther 2012;14:R52. [13] Anderson JJ, Baron G, van der Heijde D, et al. Ankylosing spondylitis assessment group preliminary definition of short-term improvement in ankylosing spondylitis.Arthritis Rheum 2001;44:1876-86. 9 [14] Braun J, Davis J, Dougados M, et al. First update of the international ASAS consensus statement for the use of anti-TNF agents in patients with ankylosing spondylitis. Ann Rheum Dis 2006;65:316-20. 10 [15] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria.Arthritis Rheum 1984;27:361-8. [16] Garrett S, Jenkinson T, Kennedy LG, et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994;21:2286-91.

104 105 CHAPTER 8

Adalimumab significantly reduces the recurrence rate of anterior uveitis in patients with ankylosing spondylitis.

J C van Denderen1 I M Visman1 M T Nurmohamed1,2 M S A Suttorp-Schulten3 I E van der Horst-Bruinsma2

Journal of Rheumatology 2014;41:1843-8.

1 Department of Rheumatology, Jan van Breemen Research Institute | Reade, Amsterdam. 2 Department of Rheumatology, VU University Medical Center, Amsterdam. 3 Department of , Onze Lieve Vrouwe Gasthuis, Amsterdam. Chapter 8 Anterior uveitis

ABSTRACT INTRODUCTION

Objective Ankylosing spondylitis (AS) is a chronic inflammatory disease that starts at a young To investigate whether use of adalimumab decreases the frequency of attacks of age and causes inflammatory back pain, stiffness, and bone deformation of the anterior uveitis (AU) in patients with ankylosing spondylitis (AS). sacroiliac joints and vertebral spine. Recently, treatment with tumor necrosis factor 1 (TNF)-blocking agents, such as infliximab, etanercept, adalimumab, and golimumab, Methods has proven to be very effective in treatment of spinal inflammation in AS [1​-4]. Consecutive patients with AS, visiting an outpatient clinic and treated for at least 12 weeks with adalimumab, were enrolled. The number of attacks of AU in the In AS, in addition to the spine, other organs can be affected, like the eye (anterior 2 year before start and during treatment were assessed by patient history and uveitis), gut (inflammatory bowel disease), and skin (psoriasis). ophthalmological controls. Anterior uveitis (AU) is strongly associated with the HLA-B27 antigen. The 3 Results occurrence of AU is increased in the HLA-B27-positive population, with a lifetime In the 77 patients a total of 52 AU attacks occurred in the year before baseline cumulative incidence of 1% compared with 0.2% in the general population [5].​ (68 attacks per 100 patient-yrs), whereas during adalimumab treatment 19 attacks Up to 47% of patients with AU are positive for the HLA-B27 antigen [6]. The mean were seen (14 per 100 patient-yrs; reduction rate 80%). Twenty-six patients with prevalence of AU in AS was 33% in a systematic literature review, and increased 4 AU in the year before start of adalimumab treatment had recurrent attacks, with with disease duration [7]​. More importantly, AU can be the first presenting a median number of 2.0 AU attacks per year [interquartile range (IQR) 1.00–3.00], symptom of AS [8-10]. The attacks of uveitis are usually unilateral and recurrent whereas during treatment this decreased to 10 patients with a median number and cause sudden ocular pain, with redness and . These attacks might of 0.56 attacks per year (IQR 0.30–0.75). Hence, the number of attacks per year lead to inflammatory debris accumulating in the anterior chamber, which may 5 decreased by 72% (p = 0.000). cause pupillary and lens dysfunction and blurring of vision. Most of the time the uveitis subsides in several weeks, but in some cases glaucoma and severe visual Conclusion impairment may occur if adequate treatment is delayed [11]​. Acute attacks call 6 In patients with AS, a significant reduction in the number of AU attacks, as well as in for urgent treatment by the ophthalmologist with topical (or in some cases even the number of attacks per patient, was observed during adalimumab treatment. systemic) corticosteroids.

In refractory uveitis (often panuveitis, idiopathic or secondary to an autoimmune 7 disease), infliximab proved to be beneficial, but overall the degree of efficacy of the different TNF-blocking agents for extraspinal manifestations in AS, such as acute AU, varies [12-15]​. Data available at the start of the current study showed a different 8 pattern for infliximab compared with etanercept: etanercept was associated with a smaller reduction in the recurrence rate of the attacks in comparison to infliximab [16-19]​. However, the majority of these studies in AS were based on retrospective analyses of clinical trials or open-label studies. Moreover, registrations of the attacks 9 were reported by the patients and ophthalmologic controls were not systematically performed. 10 Therefore, our objective was to examine whether the use of adalimumab decreases the frequency of attacks of AU in patients with AS, who received this treatment for their activity. This prospective follow-up study was carried out in close collaboration with an ophthalmologist who evaluated the patients at several times during the investigation.

108 109 Chapter 8 Anterior uveitis

MATERIALS AND METHODS measures were the changes in disease variables and the efficacy of adalimumab on disease activity of AS measured with the ASAS 20%, ASAS 40%, and BASDAI 50% Consecutive patients with AS (who fulfilled the modified New York criteria, attended responses during treatment [27]​. the outpatient clinic of the Jan van Breemen Institute Reade, and fulfilled the Assessment in SpondyloArthritis international Society (ASAS) criteria for treatment Statistical analysis 1 with a TNF-blocking agent were enrolled [20, 21]​. They were treated for at least The distribution of continuous variables was tested for normality. Data are 12 weeks with 40 mg adalimumab every other week, while other medication represented as mean ± standard deviation (SD), median, and interquartile range was continued. The number of attacks of AU in the year before start and during (IQR) or percentages. A paired t-test, or when appropriate the Wilcoxon signed- adalimumab treatment was assessed by patient history and ophthalmological ranks test, was used to determine significant changes from baseline; missing values 2 controls at baseline and yearly thereafter. The study started August 2006 and follow- were excluded in listwise fashion. Binary variables (such as occurrence of uveitis) up ended January 1, 2012, or upon discontinuation of adalimumab treatment for any were analyzed with the McNemar test. P values < 0.05 were considered statistically reason. All patients gave written informed consent prior to enrolment, and the local significant. All analyses were performed using SPSS version 17.0. 3 medical ethical committee gave its approval for this study. RESULTS Historical data were collected by patient questions on the occurrence of uveitis in the past, and confirmative answers were verified with the medical records of the A total of 100 consecutive patients with AS were screened for study, of whom 90 4 treating ophthalmologist. Additionally, an examination by our ophthalmologist was were included and 10 were not: 5 patients never started with adalimumab therapy performed at baseline and every 12 months during the first 2 years of treatment. In for several reasons, 4 declined to participate in the study and 1 declined due to case of a uveitis attack the patient was treated by his or her own ophthalmologist, pregnancy. Out of these 90 patients, 13 were excluded: 2 withdrew consent and who was always contacted by our study ophthalmologist afterwards. The number of 11 patients had incomplete follow-up (Figure 1). The remaining 77 patients were 5 attacks of uveitis during treatment with adalimumab was compared with the number included in the final analyses. Demographic and disease characteristics at baseline of attacks in the year before therapy. According to the Standardization of Uveitis are shown in Table 1. Disease-modifying antirheumatic drugs were used at baseline Nomenclature (SUN) working group a relapse in < 3 months after discontinuing by 27% of patients (e.g., sulfasalazine or methotrexate) and 21% did not respond to 6 treatment was defined as chronic uveitis; therefore a maximum of 3 flares per year prior anti-TNF treatment (etanercept and in 1 case infliximab; Table 1). The patients were counted as separate episodes [22]. Patients with insufficient data (no previous were predominantly male (61%) with a median disease duration of 7 years. The uveitis data or follow-up data available, or less than 12 weeks follow-up) were prevalence of HLA-B27 antigen was 76%, and 33 patients (43%) had experienced at excluded from the analysis. least 1 attack of uveitis. 7

The patients were assessed by a physician and research nurse at regular intervals. Comparison between patients with and those without uveitis showed that among The following measurements and questionnaire data were collected: swollen joint patients who had experienced at least 1 attack of uveitis, the prevalence of the 8 count, spinal measurements [Bath Ankylosing Spondylitis Metrology Index (BASMI), HLA-B27 antigen was significantly higher: 29/33 (94%), compared with those who Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing never had uveitis: 28/44 (64%) (p = 0.003). The median BASMI was lower in the Spondylitis Functional Index (BASFI), Bath Ankylosing Spondylitis Global score patients with uveitis-ever (score 1.0; IQR 1.0–2.5) compared to the rest of the patients (BASG), and ASQOL (AS Quality of Life questionnaire)], and the occurrence of (score 3.0; IQR 1.5–5.0) (p = 0.007). The median disease duration of patients with 9 extraspinal manifestations (like colitis, psoriasis, and uveitis) [23​-26]. In addition, uveitis-ever (12.0 yrs; IQR 2.8–19.8) was longer than that in patients who never had laboratory tests were performed, e.g., erythrocyte sedimentation rate (ESR) and uveitis (7.0 yrs; IQR 1.5–12.5), but this difference was not significant (p = 0.069). No C-reactive protein (CRP). Visits took place at screening, baseline (within 2 weeks after other significant differences were seen between patients with and those without 10 screening), after 1, 3, 6, 9, and 12 months during the first year, and every 6 months uveitis-ever in other patient or disease characteristics (data not shown). thereafter.

The primary outcome measure was the occurrence of attacks of AU in the year before baseline and during treatment with adalimumab. The secondary outcome

110 111 Chapter 8 Anterior uveitis

Table 1. Demographic and disease characteristics of patients with ankylosing spondylitis (AS) at baseline (n = 77) and during follow-up with adalimumab treatment. Mean values (standard deviation) or median (interquartile range, IQR) are shown, as appropriate.

Characteristic Baseline 12 Weeks 52 Weeks Male, n (%) 47 (61) 1 Age, yrs, mean (SD) 45 (10) White, n (%) 53 (75) Duration of inflammatory back pain, median 17 (10–25) yrs (IQR) 2 Disease duration (diagnosis), median yrs 7 (2–15) (IQR) Years since first uveitis, median (IQR) 11 (4–23) 3 Anterior uveitis ever, n (%) 33 (43) Anterior uveitis last year, n (%) 26 (34) HLA-B27 present, n (%) (n = 75) 57 (76) Psoriasis, n (%) 5 (6) 4 Inflammatory bowel disease, n (%) 9 (12) Use of NSAID at baseline, n (%) 65 (84) Use of DMARD at baseline, n (%)** 21 (27) 5 Use of prednisone, n (%) 4 (5) Previous use of other anti-TNF therapy*** 16 (21) BASDAI, mean (SD) 5.5 (2.1) 3.8 (2.5)* 3.7 (2.4)* Figure 1. A total of 100 consecutive patients with ankylosing spondylitis (AS) were screened for BASFI, median (IQR) 4.7 (2.7–6.6) 3.6 (1.1–5.4)* 2.9 (0.9–5.6)* study; 90 were included and 10 were not. 6 BASG, mean (SD) 6.6 (2.1) 4.9 (2.4)* 4.0 (2.5)* BASMI, median (IQR) 2.0 (1.0–4.0) 2.0 (1.0–3.0) 2.0 (1.0–3.0) In total, 67 of the 77 patients (87%) were seen by the ophthalmologist at baseline ASQOL, median (IQR) 9.0 (5.8–13.0) 6.5 (2.0–10.0)* 4.0 (2.0–9.0)* 7 and 44 (57%) during follow-up; the other data were retrieved from hospital charts No. patients with at least 1 swollen joint 18 (25) 6 (9)* 3 (6)* (SJC-72) (%) and protocol visits to the research physician. The median follow-up period (i.e., ESR, median mm/h (IQR) 14 (7–31) 6 (3–15)* 6 (2–14)* duration of treatment) was 1.74 years (IQR 0.83–2.84), with a total of 139 patient- years. Three subjects had symptomatic uveitis at baseline. Ophthalmological 8 CRP, median mg/l (IQR) 4 (2–13) 1 (1–4)* 1 (1–3)* examinations in patients without symptoms did not reveal any uveitis. No. ASAS-20% responders (%) 25 (56) 20 (53) No. ASAS-40% responders (%) 1 (2) 3 (8) Out of the 77 patients at baseline, 51 (66%) had no attacks of uveitis in the year No. BASDAI-50% responders (%) 31 (55) 19 (41) 9 before treatment, nor during treatment. No patient developed uveitis for the first * p < 0.05 followup towards baseline. ** DMARD: In most cases sulfasalazine (n = 10) or time during adalimumab treatment. Seven patients (9%) were classified as having methotrexate (n = 8). *** Previous use of other anti-TNF agent: etanercept (n = 15), infliximab chronic uveitis, but during adalimumab treatment chronic uveitis did not occur. (n = 1). NSAID: nonsteroidal antiinflammatory drug ; DMARD: disease-modifying drug; 10 BASDAI: Bath AS Disease Activity Index (0–10); BASFI: Bath AS Functional Index (0–10); BAS-G, Bath AS Global Score (0–10); BASMI, Bath AS Metrology Index (0–10); ASQOL, AS Quality of In all patients, the total number of uveitis attacks was 52 in the year before baseline Life (0–18); SJC: swollen joint count; ESR: erythrocyte sedimentation rate; CRP: C-reactive (i.e., 68 attacks per 100 patient-yrs), whereas during adalimumab treatment 19 protein; ASAS: Assessment of Spondyloarthritis International Society; TNF: tumor necrosis attacks were seen (14 per 100 patient-yrs; reduction rate 80%). Uveitis attacks were factor. treated with topical steroids.

112 113 Chapter 8 Anterior uveitis

Among the 26 (34%) patients with uveitis in the year before treatment, 10 had flares Table 2. Number of anterior uveitis (AU) attacks/100 patient-years and reduction during anti- of uveitis during treatment and 16 had no uveitis (Figure 2). This indicates a 62% tumor necrosis factor (anti-TNF) therapy compared to placebo, i.e., year before therapy in this decrease in the number of patients with uveitis attacks (p < 0.0001). The patients with study compared with other studies. Duration of followup in patient-years and mean patient- years/patient. chronic uveitis responded as well as the other uveitis patients: 27% of the patients with uveitis in the year before baseline had chronic uveitis and 2 of these had uveitis Study No. AU Attacks/ No. AU Attacks/ Reduction, Followup, Mean 100 Patient-yrs 100 Patient-yrs % Patient-yrs Patient-yrs/ 1 after start of adalimumab (20% of all patients with uveitis during therapy). During Placebo During Anti-TNF (No. patients) Patient Braun, 2005 [1]* 16 ETN 8 49 430 (297) 1.4 IFX 3 78 146 (90) 1.6 2 Sieper, 2010 [28]* 19 ETN 12 38 1137 (1074) 1.1 Before anti-TNF Rudwaleit, 2009 [30] 15 ADA 7 51 363 (1250) 0.3 Current study 68 ADA 14 80 139 (77) 1.8 3

*randomized controlled trial (RCT) + open-label (OL) phase; ETN: etanercept; IFX: infliximab; ADA: adalimumab. 4

Discontinuation of therapy and adverse events Twenty-eight (36%) of the patients stopped adalimumab therapy, usually because of inefficacy (17 patients, 61% of dropouts) or adverse events (3 patients, 11%). 5 Other reasons for dropping out were as follows: moving out of area (2), withdrawal of informed consent (1), and other reasons (5). No patient discontinued because of Figure 2. Among 26 patients with uveitis in the year before treatment, 10 had flares during an AU episode. The median follow-up period of the dropouts was 0.82 years (IQR 6 treatment and 16 had no uveitis. *p < 0.0001. 0.37–1.48).

A total of 268 adverse events (AE), mostly mild, were observed among 67 (87%) The 26 patients with uveitis in the year before start of adalimumab experienced patients, of whom 43 (64%) were still on drug and 24 (36%) had stopped. Infection 7 recurrent attacks, with a median number of 2.0 uveitis attacks per year (IQR 1.00– was the most common AE observed (69 times). One patient was hospitalized (due to 3.00), whereas during treatment this dropped to 10 patients, with a median number local skin infection), and in 13 cases antibiotics were given. No patient died during of 0.56 attacks per year (IQR 0.30–0.75). Hence, the number of attacks per year adalimumab treatment. 8 dropped by 72% (p < 0.0001). DISCUSSION The 26 patients with uveitis in the year before therapy had 200 flares per 100 patient- years before therapy and during adalimumab had 19 flares in 61 patient-years The number of attacks of uveitis in patients with AS before and during adalimumab 9 (31 flares per 100 patient-years; reduction rate 84%; Table 2). treatment revealed a significant decrease in the number of patients with uveitis (62%) during treatment, as well as a decrease in the number of attacks per patient Disease measures at follow-up are shown in Table 1. In general, improvement (72%), even in patients with chronic uveitis. No patient developed uveitis for the first 10 was seen in measures for disease activity, functionality, and quality of life. These time during adalimumab treatment. The majority (87%) of patients remained free of improvements were statistically significant, except for the BASMI. The percentages uveitis attacks for the entire follow-up period. A significant improvement of disease of patients with ASAS 20% and BASDAI 50% response were 56 and 55, respectively, activity (ASAS 20% response in 56% of patients) was seen as well, which is similar to at Week 12 and 53 and 41, respectively, at Week 52. At Week 52 complete datasets the results of placebo-controlled trials with anti-TNF agents [3, 4].​ were available for 47 patients.

114 115 Chapter 8 Anterior uveitis

Patient characteristics of those who experienced uveitis and those who did not have placebo in patients of other studies (which described 15 to 19 flares/100 patient-yrs). any attack did not differ, except for the prevalence of HLA-B27 antigen, which was Obviously, the absolute numbers of patients with uveitis during therapy cannot be significantly higher among the patients who had experienced at least 1 attack of compared with other studies, but the rates of reduction can. We observed a high uveitis. This observation is in accord with previous studies, which show a strong rate of reduction of uveitis, even in our patients with a high incidence of uveitis or association between uveitis and HLA-B27 [5,28]​. The lower median BASMI in the chronic uveitis. 1 patients with uveitis-ever is probably due to chance, as all other disease measures were not different. A longer disease duration in patients who had ever had uveitis In addition, we cannot exclude that the high percentage of uveitis at baseline in could be expected, but this difference did not reach statistical significance in our our study might be due to a selection bias, because of preferential prescription of patients [7]. adalimumab in cases of a history of uveitis during the period of our study. On the 2 other hand, comparing the frequency of earlier treatment with another anti-TNF In contrast with other studies in AS, information about uveitis in our study was therapy in this study (21%) with that of another study (26%), the study populations obtained at visits to the ophthalmologist at baseline and during follow-up. Because appear to be comparable [30]. the visit to the ophthalmologist could not be combined with the other protocol 3 visits, not all patients were seen by the ophthalmologist. However, it is not likely that The prevalence of uveitis in AS increases with longer disease duration [7]​. However, this had a significant influence on our results as (1) this was particularly the case in the high prevalence of uveitis recorded in our study cannot be explained by longer those who never had eye problems, and (2) in case of an attack of uveitis, patients disease duration compared to other studies. In contrast, the median disease 4 were always seen by their attending ophthalmologist and this was verified by our duration in our study was 7 years (IQR 2–15), compared to a mean duration of 9.5–11 (study) ophthalmologist. years in other studies [28, 30]​.

The rates of uveitis per 100 patient-years in our study are compared with other Interestingly, in our study no patient developed AU for the first time during 5 studies that investigated this topic in Table 2. In 2 studies the incidence of uveitis adalimumab treatment, but other studies have described that a first attack of uveitis during placebo and during anti-TNF therapy was determined from several clinical may occur during therapy with several TNF-blocking agents [30-32]. trials of etanercept and infliximab, including open-label extension studies [28,​ 29]. 6 In an open-label uncontrolled study the rate of uveitis during adalimumab treatment We observed a significant and substantial reduction of the rate of recurrence of was compared with the incidence reported during the year before treatment [30]​. In attacks of acute anterior uveitis during adalimumab treatment in patients with all studies the rate of uveitis during anti-TNF treatment was lower compared to that AS, even in patients with chronic uveitis. The majority (87%) of patients remained of the placebo period or the year before therapy. The 80% reduction in recurrence completely free of uveitis attacks for the entire follow-up period. 7 rates of uveitis during adalimumab treatment in our study was higher in comparison with the other studies, which varied from 78% for infliximab, to 38%–49% with etanercept, and 68% with adalimumab (Table 2). 8

Differences in reduction of uveitis between these studies could be explained by differences in duration of follow-up. In our study the mean number of patient-years per patient was 1.8 years versus 0.3 years in the study of Rudwaleit, et al [30]. In 9 some patients the efficacy of the anti-TNF treatment could last some months, an observation that might be overlooked in studies with shorter follow-up period. 10 Another difference is that a high percentage of our patients (n = 33, 43%) had experienced uveitis at some time before treatment, whereas in most AS studies the prevalence of AU ranged between 25% and 40% among patients [11].​ Comparing the rates of uveitis flares/100 patient-years (Table 2) also shows a higher number (n = 68) in our study compared with the lower rates before treatment or during

116 117 Chapter 8 Anterior uveitis

REFERENCES

[1] Braun J, Brandt J, Listing J, Zink A, Alten R, Golder W, et al. Treatment of active [17] Reddy AR, Backhouse OC. Does etanercept induce uveitis? Br J Ophthalmol 2003;87:925. ankylosing spondylitis with infliximab: A randomised controlled multicentre trial.Lancet [18] Rosenbaum JT. Effect of etanercept on iritis in patients with ankylosing spondylitis. 2002;359:1187–93. Arthritis Rheum 2004;50:3736–7. [2] Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosing spondylitis by inhibition of [19] Guignard S, Gossec L, Salliot C, Ruyssen-Witrand A, Luc M, Duclos M, et al. Efficacy 1 tumor necrosis factor alpha. N Engl J Med 2002;346:1349–56. of tumour necrosis factor blockers in reducing uveitis flares in patients with [3] Van der Heijde D, Kivitz A, Schiff MH, Sieper J, Dijkmans BAC, Braun J, et al; and ATLAS spondylarthropathy: A retrospective study. Ann Rheum Dis 2006;65:1631–4. Study Group. Efficacy and safety of adalimumab in patients with ankylosing spondylitis: [20] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing Results of a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis spondylitis. A proposal for modification of the New York criteria.Arthritis Rheum 2 Rheum 2006;54:2136–46. 1984;27:361–8. [4] Inman RD, Davis JC Jr., van der Heijde D, Diekman L, Sieper J, Kim SI, et al. Efficacy and [21] Braun J, Pham T, Sieper J, Davis J, van der Linden S, Dougados M, et al; and for the ASAS safety of golimumab in patients with ankylosing spondylitis: Results of a randomized, Working Group. International ASAS consensus statement for the use of anti-tumour double-blind, placebo-controlled, phase III trial. Arthritis Rheum 2008; 58:3402–12. necrosis factor agents in patients with ankylosing spondylitis. Ann Rheum Dis 2003; 3 [5] Linssen A, Rothova A, Valkenburg HA, Dekker-Saeys AJ, Luyendijk L, Kijlstra A, et al. 62:817–24. The lifetime cumulative incidence of acute anterior uveitis in a normal population and [22] Jabs DA, Nussenblatt RB, Rosenbaum JT. Standardization of Uveitis Nomenclature (SUN) its relation to ankylosing spondylitis and histocompatibility antigen HLA-B27. Invest Working Group. Standardization of uveitis nomenclature for reporting clinical data. 4 Ophthalmol Vis Sci 1991;32:2568–78. Results of the First International Workshop. Am J Ophthalmol 2005;140:509–16. [6] Wakefield D, Dunlop I, McCluskey PJ, Penny R. Uveitis: Aetiology and disease associations [23] Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mallorie P, et al. A new approach in an Australian population. Aust NZ J Ophthalmol 1986;14:181–7. to defining functional ability in ankylosing spondylitis: The development of the Bath [7] Zeboulon M, Dougados M, Gossec L. Prevalence and characteristics of uveitis in the Ankylosing Spondylitis Functional Index. J Rheumatol 1994;21:2281–5. 5 spondyloarthropathies: A systemic literature review. Ann Rheum Dis 2008;67:955–9. [24] Garrett S, Henkinson T, Kennedy G, Whitelock H, Gaisford P, Calin A. A new approach to [8] Pato E, Bañares A, Jover JA, Fernández-Gutiérrez B, Godoy F, Morado C, et al.Undiagnosed defining disease status in ankylosing spondylitis: The Bath Ankylosing Spondylitis Disease spondyloarthropathy in patients presenting with anterior uveitis. J Rheumatol Activity Index. J Rheumatol 1994;21:2286–91. 2000;27:2198–202. [25] Jenkinson TR, Mallorie PA, Whitelock HC, Kennedy LG, Garret SL, Calin A. Defining spinal 6 [9] Linder R, Hoffmann A, Brunner R. Prevalence of the spondyloarthritides in patients with mobility in ankylosing spondylitis (AS), the Bath AS Metrology Index. J Rheumatol uveitis. J Rheumatol 2004;31:2226–9. 1994;21:1694–8. [10] Monnet D, Breban M, Hudry C, Dougados M, Brézin AP. Ophthalmic findings and [26] Jones SD, Steiner A, Garrett SL, Calin A. The Bath Ankylosing Spondylitis Patient Global frequency of extraocular manifestations in patients with HLA-B27 uveitis: A study of 175 Score (BAS-G). Br J Rheumatol 1996;35:66–71. 7 cases. Ophthalmology 2004;111:802–9. [27] Van der Heijde D, Calin A, Dougados M, Khan MA, van der Linden S, Bellamy N; and on [11] Gensler L. Clinical features of ankylosing spondylitis. In: Hochberg MC, Silman AJ, Smolen behalf of the ASAS Working Group. Selection of instruments in the core set for DC-ART, JS, et al. editors. Rheumatology. Philadelphia: Mosby, 2011. SMARD, physical therapy, and clinical record keeping in ankylosing spondylitis. Progress report of the ASAS Working Group. J Rheumatol 1999;26:951–4. [12] El-Shabrawi Y, Hermann J. Anti-tumor necrosis factor-alpha therapy with infliximab as an 8 alternative to corticosteroids in the treatment of human leukocyte antigen B27-associated [28] Sieper J, Koenig A, Baumgartner S, Wishneski C, Foehl J, Vlahos B, et al. Analysis acute anterior uveitis. Ophthalmology 2002;109:2342–6. of uveitis rates across all etanercept ankylosing spondylitis trials. Ann Rheum Dis 2010;69:226–9. [13] Kruithof E, Kestelyn P, Elewaut C, Elewaut D, van den Bosch F, Mielants H, et al. Successful use of infliximab in a patient with treatment resistant spondyloarthropathy related uveitis. [29] Braun J, Baraliakos X, Listing J, Sieper J. Decreased incidence of anterior uveitis in 9 Ann Rheum Dis 2002;61:470. patients with ankylosing spondylitis treated with the anti-tumor necrosis factor agents infliximab and etanercept. Arthritis Rheum 2005;52:2447–51. [14] Murphy CC, Ayliffe WH, Booth A, Makanjuola D, Andrews PA, Jayne D. Tumor necrosis factor alpha blockade with infliximab for refractory uveitis and scleritis. Ophthalmology [30] Rudwaleit M, Rødevand E, Holck P, Vanhoof J, Kron M, Kary S, et al. Adalimumab 10 2004;111:352–6. effectively reduces the rate of anterior uveitis flares in patients with active ankylosing spondylitis: Results of a prospective open-label study. Ann Rheum Dis 2009;68:696–701. [15] Suhler EB, Smith JR, Wertheim MS, Lauer AK, Kurz DE, Pickard TD, et al. A prospective trial of infliximab therapy for refractory uveitis. Arch Ophthalmol 2005;123:903–12. [31] Tiliakos AN, Tiliakos NA. Ocular inflammatory disease in patients with RA taking etanercept: Is discontinuation of etanercept necessary? J Rheumatol 2003;30:2727. [16] Foster CS, Tufail F, Waheed NK, Chu D, Miserocchi E, Baltatzis S, et al. Efficacy of etanercept in preventing relapse of uveitis controlled by methotrexate. Arch Ophthalmol [32] Wendling D, Paccou J, Berthelot JM, Flipo RM, Guillaume-Czitrom S, Prati C, et al. New 2003;121:437–40. onset of uveitis during anti-tumor necrosis factor treatment for rheumatic diseases. Semin Arthritis Rheum 2011;41:503–10. 118 119 CHAPTER 9

Etanercept increases bone mineral density in ankylosing spondylitis, but does not prevent vertebral fractures.

M A C van der Weijden1, 2 J C van Denderen2 W F Lems1, 2 M T Nurmohamed1, 2 B A C Dijkmans1 I E van der Horst-Bruinsma1, 2

Submitted

1 VU University Medical Center, Department of Rheumatology, Amsterdam. 2Jan van Breemen Research Institute | Reade, Department of Rheumatology, Amsterdam. Chapter 9 Bone mineral density and vertebral fractures

ABSTRACT INTRODUCTION

Objective Ankylosing spondylitis (AS) is characterized by chronic inflammation leading to Ankylosing spondylitis (AS) is characterized by chronic inflammation leading to ankylosis of the spine and sacroiliac joints. Bone loss is a well known complication ankylosis, but also to low bone mineral density (BMD) and vertebral fractures (VF). of AS [1, 2], which is highly prevalent after long disease duration, but starts already 1 Treatment with TNF blockers decreases inflammation and has shown to be effective at an early stage [3, 4]. Bone loss and inflammation are probably responsible for the in increasing BMD. We studied the effects of etanercept on BMD and VF in AS occurrence of vertebral fractures (VF) in this patient group [5-7]. The pathogenesis of patients after 2 years treatment. Furthermore we studied changes in bone-markers the decrease in bone mineral density (BMD) is complex. Persistent inflammation (by (CTXI, CTXII, RANKL, OPG, Osteocalcin) and radiological damage. inflammatory cytokines like TNF-α) might be an important etiologic factor [8, 9]. 2

Methods A way to decrease inflammation is treatment with TNF-α blockers. In rheumatoid Patients with active AS, treated with etanercept for 2 years, were included. BMD arthritis, several studies showed an increase of BMD, additionally to the positive 3 lumbar spine and hips was measured at baseline and after 2 years, as well as the effects on disease activity and radiographic progression [10-12]. In AS, many studies radiological damage (mSASSS, including thoracic spine), VF (Genant method) and showed positive effects of TNF-α blockers on inflammation and disease activity, change in bone-markers. although effects on decreasing radiographic progression are disappointing [13, 14]. Considering effects of TNF-α blockers on BMD, patients treated with infliximab 4 Results showed significant increases in BMD scores over 2 years [15]. A very small study Forty-nine AS patients were included. After 2 years of etanercept, BMD hip raised (n=10) of Marzo-Ortega et al. showed that etanercept increased BMD in a short with 2.2% (p=0.014) and BMD lumbar spine with 7.0% (p<0.001). The BASDAI follow-up study of 6 months [16]. Arends et al. also showed an increase in BMD and, decreased significantly (p<0.001) as well as CRP and ESR (p<0.001). Despite in addition, an effect in favour of bone formation by measuring bone-markers in 5 etanercept therapy, the number of patients with VF more than doubled (from 6 to 15 patients treated with different types of TNF-α blockers [17], but whether there were patients, p=0.004). Also the radiological damage increased significantly over time differences between the effects of TNF-α blockers did become clear. Furthermore, (from 12.1 to 18.5, p<0.001), however no significant change in bone-markers was clinically relevant outcome measures like VF, radiographic progression and disease 6 found. activity combined in one study were not performed.

Conclusion Therefore, the aim of this study was to measure the effects of 2 years of etanercept This prospective longitudinal observational cohort study showed that after 2 on bone quality by measuring change of BMD and the incidence of VF. Further, we 7 years etanercept, BMD of hip and spine increased significantly, but the number of assessed changes in bone-markers and the effects on radiographic damage. patients with VF and the severity of VF increased as well. Besides that, radiological progression, including the thoracic spine, increased significantly. Thus, the PATIENTS AND METHODS 8 favourable bone preserving effect is accompanied by unfavourable outcomes on VF and radiological damage. Study population AS patients, who fulfilled the modified New York criteria for AS and were eligible for treatment with anti-TNF-α (etanercept) according to the ASAS guidelines, were 9 recruited from the Jan van Breemen Research Institute/Reade, a large outpatient rheumatology center in Amsterdam [18]. The data for this open prospective follow- up study were collected systematically every 3 months during the first year and 10 twice yearly thereafter.

AS patients were treated during 2 years with etanercept as decided by their physician with etanercept (25 mg twice a week or 50 mg once a week) if they had previously failed on at least two nonsteroidal anti-inflammatory drugs (NSAIDs)

122 123 Chapter 9 Bone mineral density and vertebral fractures

and if they had active disease (Bath Ankylosing Spondylitis Disease Activity Index in serum using commercial assays according to the instructions of the manufacturer (BASDAI) ≥ 4). (Roche Diagnostics, Mannheim, Germany). CTX-II was determined using a urine Demographic data, such as HLA-B27 status, extra-articular manifestations, current Cartilaps ELISA (from Immunodiagnostic Systems; IDS) for the quantification use of medication, including anti-osteoporotic drugs, as well as data on known risk of degradation products of C-terminal telopeptides of type II collagen in human factors for osteoporosis, such as age, sex, race, smoking and disease duration urine. Levels of osteoclast-regulating proteins, including total RANKL and OPG, 1 [19, 20], were collected at baseline. were determined in serum using an ELISA (from Immundiagnostik AG, Bensheim, Germany). Bone formation was measured by osteocalcin, using commercial assays The protocol was approved by the local Medical Ethics Committee and all patients according to the instructions of the manufacturer (Roche Diagnostics, Mannheim, provided written informed consent. Germany). All assays on the analyzer had an intra-assay and inter-assay coefficient 2 of variation of ≤ 5%. The ELISA’s had an intra-assay and inter-assay coefficient of Outcome measures variation of ≤ 10%.

Bone mineral density Radiographic progression 3 First, the change in BMD of the lumbar spine (L2-L4) and left proximal femur after The degree of radiological damage of the spine, before and after treatment with 2 years of etanercept was investigated. Each patient was measured by dual energy etanercept, was also investigated. The modified Stoke Ankylosing Spondylitis Spine X-ray absorptiometry (DEXA) using Lunar (Lunar expert DPX-IQ, Oldelft). Results Score (mSASSS) was used as the index of radiological damage of the cervical and 4 were presented as BMD (g/cm2), T-scores and Z-scores. The T-score corresponds to lumbar spine [23]. The lateral radiographs of the spine of each patient, at baseline the number of standard deviations (SD) from the normal mean obtained from young and after 24 months, were examined by two experienced investigators (CvD and healthy adults and the Z-score is the T-score with a correction for age. IvdH). Additionally, the radiographs of the thoracic spine were assessed as well and osteoporosis are defined according to the World Health Organization (WHO): 1) (T9-T12), although they are not implemented in the official scoring method of the 5 osteoporosis (T-score ≤ -2.5 in spine and/or hip), 2) osteopenia (-2.5 < T-score < -1.0 in mSASSS. However, since the low thoracic spine might have an additive effect on the spine and/or hip, without osteoporosis) [21]. sensitivity to change [24] and is a predilection place for VF, the thoracic X-rays were included in the measurement. 6 Vertebral fractures Further, we investigated the occurrence of VF. Radiographs of thoracic and lumbar Disease activity spine were made at baseline and after 24 months. The lateral radiographs were Disease activity measures included the disease activity score BASDAI [25], whereby evaluated chronologically for VF by two experienced investigators (WL and BD), the functional capacity scores BASFI (Bath Ankylosing Spondylitis Functional Index) 7 who were blinded for medication the patients received. Vertebral deformities were [26] and BASMI (Bath Ankylosing Spondylitis Metrology Index) [27] measured the determined by grading each vertebral body (T4-L5) according to the Genant criteria physical function. The ASAS criteria for response were applied to define response for fractures [22]. In Genant’s semiquantitative assessment, the vertebrae receive [28] as a 50% improvement or as an absolute improvement of 2 points of the 8 a severity grade based on the visually apparent degree of vertebral height loss. BASDAI (0-10 scale) and an expert opinion in favour of continuation of treatment The reduction in height is divided in grades on a scale of 0-3: grade 0 (normal) after 3 months. represents a reduction in anterior, middle, and/or posterior vertebral heights of less than 20%, grade 1 (mild) represents a reduction of 20-25%, grade 2 (moderate) Statistical analysis 9 reduction of 25-40% and grade 3 (severe) more than 40% reduction. VF were defined Categorical variables were calculated as frequencies and percentages. Continuous as a reduction of ≥ 20% of the vertebral body height [22]. variables were reported as mean and SD, or when skewed, as median and IQR. To examine the longitudinal changes in BMD (and T- and Z-scores), BMD was first 10 Markers of bone turnover tested for normality (with the Shapiro-Wilks test), and subsequently the paired Different bone-markers, CTX-I, CTX-II, RANKL, OPG and osteocalcin were obtained t-test or Wilcoxon’s signed rank test was used. Differences in VF was tested with at 0, 3, 6 and 12 months. Non-fasting serum and urine were collected and stored at the McNemar test. Bone-markers were tested for a linear trend with regression -20°C until analyses. Bone resorption was measured by CTX-I and CTX-II. CTX-I was analyses. To detect differences between different time moments the Friedman test determined by β-isomerised carboxy terminal telopeptide of type 1 collagen (β-CTx) was used. Radiological damage was analysed first by testing for normality and

124 125 Chapter 9 Bone mineral density and vertebral fractures

subsequently the change over time was tested with the non-parametric Wilcoxon’s BMD variables signed rank test. Disease activity changes in BASDAI, CRP and ESR were also sBMD hipb 0.903 (0.152) analysed with the non-parametric Wilcoxon’s signed rank test because of the skewed sBMD L2-L4b 1.141 (0.203) distribution. T-score hipb -0.92 (1.14) b Statistical analyses were performed with SPSS statistical software, version 20.0 T-score L2-L4 -0.29 (1.77) Z-score hipb -0.81 (1.04) 1 (SPSS, Chicago, IL). P-values less than 0.05 were considered as significant. Z-score L2-L4b -0.31 (1.71) Medication RESULTS NSAIDs (current)a 49 (100) DMARDs (current)a 12 (24.5) 2 Patient characteristics Corticosteroids (current)a 1 (2.0) In total, 49 patients with AS were enrolled and monitored after starting with Bisphosphonates (current)a 3 (6.1) etanercept. The mean follow-up duration of these patients was 2.3 years. The anumber (%), bmean (SD), cmedian (IQR) baseline demographics and clinical features are shown in Table 1. ESR=Erythrocyte Sedimentation Rate, [normal value]; CRP=C-Reactive Protein, [normal 3 Most patients were male (82%), the mean age was 42 years and the mean disease value]; BASDAI=Bath Ankylosing Spondylitis Disease Activity Index; BASFI=Bath Ankylosing Spondylitis Functional Index, BASMI=Bath Ankylosing Spondylitis Metrology Index; BMI= Body duration was 12.2 years. Three patients had a total hip replacement at inclusion. Mass Index; sBMD=standardized Bone Mineral Density in g/cm2. There was a high disease activity before start of therapy and the majority responded 4 well.

Bone mineral density All 49 patients had two DEXA scans of the spine and 46 patients had a DEXA of the Table 1. Baseline characteristics AS-etanercept cohort (N=49). 5 hip (three patients had a total hip replacement). At baseline, 12% of the patients Demographic variables had osteoporosis, 45% osteopenia and 43% had a normal BMD. After 2 years of Mena 40 (81.6) etanercept this changed to 4% osteoporosis, 41% osteopenia and 55% normal BMD Age (years)b 41.8 (9.2) 6 Caucasian racea 38 (77.6) (Table 2). Disease related variables After 2 years of treatment, BMD hip raised significantly with 2.2% (5.7; p=0.014) Disease duration (years)b 12.2 (9.1) and BMD lumbar spine raised significantly with 7.0% (9.5; p<0.001). The mean T- and Symptom duration (years)c 15.8 (9.9-23.4) Z-scores showed the same significant increase of the hip (p=0.037 vs p=0.002) and b 7 Follow-up duration (years) 2.3 (0.7) lumbar spine (both p<0.001). HLA-B27 positivitya 43 (87.8) ESR (mm/hr), [<20]c 20.0 (6.0-39.0) CRP (mg/l), [<10]c 14.0 (3.0-39.0) Table 2. Bone mineral density of femur and spine. BASDAI (0-10)b 5.7 (1.6) 8 BASFI (0-10)b 5.7 (2.1) t = 0 year, before etanercept t = 2 years, after etanercept BASMI (0-10)b 4.4 (2.3) Hip* Spine Total BMD Hip* Spine Total BMD Uveitis (history of)a 17 (34.7) Normal BMDa 23 (46.9) 29 (59.2) 21 (42.9) 27 (55.1) 36 (73.5) 27 (55.1) Psoriasis (history of)a 6 (12.2) 9 Osteopeniaa 19 (38.8) 16 (32.7) 22 (44.9) 17 (34.7) 12 (25.5) 20 (40.8) Inflammatory bowel disease (history of)a 3 (6.1) a Peripheral arthritis (history of)a 16 (32.7) Osteoporosis 4 (8.2) 6 (12.2) 6 (12.2) 2 (4.1) 1 (2) 2 (4.1) Radiographic damage Normal BMD= T-score ≥ -1.0, Osteopenia= -2.5 < T-score < -1.0, Osteoporosis= T-score ≤ -2.5. 10 c Total mSASSS (0-72) 10.0 (3.8-35.5) anumber (%), *calculated in 46 patients, because 3 patients had a hip replacement. Total mSASSS+ThSpine (0-90)c 12.1 (6.8-42.7) BMD related variables BMI (kg/m2)b 26.3 (3.4) Post-menopausal statusa 2 (4.1) One or more prevalent vertebral fracturesa 6 (12.2) Smoking (current)a 24 (49.0)

126 127 Chapter 9 Bone mineral density and vertebral fractures

Vertebral fractures Figure 1. Distribution of bone turnover markers during treatment with etanercept. At baseline six patients (12.2%) already had at least one VF. After 2 years of etanercept, the number of patients with one or more VF was more than doubled to 15 patients (30.6%) (p=0.004). Most VF were localised in the (mid)thoracic spine. Not only the number of patients with VF increased significantly over 2 years, but also the 1 severity (grade) of the VF, from 4 fractures (out of 8) graded 2 or more to 13 fractures (out of 21) graded at least 2 (Table 3). Analyses for risk factors for the development of VF did not show any parameter to be associated with these incident VF (such as age, BMD, disease activity, radiological damage; data not shown). 2

Table 3. Vertebral fractures before and after etanercept in AS (N=49). 3 Vertebral fractures t=0 year, before etanercept t=2 years, after etanercept N patients Number of VF N patients Number of VF 0 Vertebral fracturea 43 (87.8) 34 (69.4)

1 Vertebral fracturea 4 (8.2) 9 (18.4) 4 Grade I 2 5 Grade II 2 4 Grade III 0 0 5 2 Vertebral fracturesa 2 (4.1) 6 (12.2) Grade I 2 3 Grade II 2 7 Grade III 0 2 6

Total number of VF 8 21 Total patients with VFa 6 (12.2) 15 (30.6) 7 anumber of patients(%) VF= Vertebral Fracture, Grade I= reduction of vertebral height 20-25%, Grade II= reduction of vertebral height 25-40%, Grade III= reduction of vertebral height >40%. 8 Markers of bone metabolism Boxplots of the distribution of bone-markers over time, as well as the change in disease activity (BASDAI) and inflammation (CRP) are shown in Figure 1. Bone- markers were tested for linear trend with regression analyses, but there was 9 Vertical scale left: scale of different bone-markers. Vertical scale right: scale of BASDAI and CRP. no significant trend over time for the bone ’resorption’ markers, the ‘osteoclast- Horizontal: different time moments; 1= 0 month, 2= 3 months, 3= 6 months, 4= 12 months. regulation’ markers, nor for the bone ‘formation’ marker. To detect differences Boxplots= different bone-markers, Δ= CRP, □= BASDAI. between different time moments, the Friedman test was used. No significant 10 changes in bone-markers were detected over time, except for OPG (which showed a Radiological progression decreasing trend). However the RANKL/OPG ratio did not change significantly over The median radiological damage (total mSASS-score) at baseline was 10.0 (3.8-35.5) time. and increased after 2 years to 15.5 (5.5-42.5). The total mSASSS+ThSpine had a median at baseline of 12.1 (6.8-42.7) and progressed to 18.5 (8.7-52.0). There was a significant difference between the radiological damage at baseline and after 2 years

128 129 Chapter 9 Bone mineral density and vertebral fractures

(p<0.001) measured by as well the mSASSS as the mSASSS+ThSpine. Correlation inflammation through etanercept therapy and despite the increase of BMD. The tests did not show a significant relation between BMD change over time and the interpretation of VF in AS is a challenge, since the method of Genant does not radiological progression. Further, there was neither a correlation between bone- differentiate between AS-related deformities, degenerative changes or osteoporotic markers and radiological progression, nor between disease activity parameters and fractures. The presence of low BMD and the localization of the VF suggest that radiological progression. these fractures are ‘real’ osteoporotic fractures. We have documented earlier in a 1 cohort of early Spondylarthropathy patients a high number of patients (15%) with Disease activity VF in especially the thoracic spine [7]. Unfortunately, VF in AS are often missed Disease activity measured by BASDAI decreased from 5.8 (5.1-6.8) at baseline to in clinical routine procedures, however diagnosing these fractures is important 2.1 (1.0-4.1) after 12 months and to 2.8 (1.0-4.4) after 24 months (p<0.001 at all time because the knowledge of existing fractures is necessary for optimal assessment of 2 points compared with baseline). According to the ASAS clinical response criteria, risk for future fractures and treatment [6, 33]. However, until now there are no clear 85% of the patients responded to etanercept at 3 months, 69% at 12 months and guidelines for how we should treat these patients. In this study the treatment of the 63% at 24 months. osteopenic/osteoporotic patients was performed by the treating rheumatologist. Disease activity measured by inflammation markers like CRP and ESR also Most patients were treated with calcium/vitD, no bisphosphonate treatment was 3 decreased significantly (p<0.001 at all time points compared with baseline). started in the period of the study. CRP decreased from 14.0 (3.0-39.0) at baseline to 2.0 (1.0-6.0) at 12 months. ESR decreased from 20.0 (6.0-39.5) at baseline to 5.0 (2.0-9.0) at 12 months. ESR and CRP The increased prevalence of VF despite the increase of BMD suggests that it is more 4 changes were strongly correlated (p<0.001). likely that despite the increase in ‘quantity’ of bone mass, the problem in AS is more There was a significant relation between the change in BMD hip and spine and due to a decrease in ‘bone quality’. A specific definition of the quality of bone is the change in inflammation markers (i.e. ESR and CRP) over 12 months. Both the difficult to be given, because multiple factors contribute to the structural integrity decrease in ESR and CRP was significantly associated with increase in BMD hip and of bone: not only the total bone mass, but also bone geometry and properties of 5 spine (ΔBMD hip: p=0.040 vs p=0.005 and ΔBMD spine: p=0.012 vs p<0.001). constituent tissue [34]. As BMD has been shown to be a limited predictor of fracture risk [35], now more clinical interest is needed for complementary measures of bone DISCUSSION quality that could improve fracture risk prediction [36]. One of these measures 6 could have been bone-markers, but unfortunately we did not find a linear trend of This prospective observational cohort study in patients with active AS demonstrated bone ’resorption’ markers (CTX-I, CTX-II), ‘osteoclast-regulating’ markers (RANKL, that after 2 years of TNF-α blocking therapy with etanercept, BMD of the hip as well OPG), or of the bone ‘formation’ marker (osteocalcin) over time during etanercept as BMD of the spine increased significantly, whereas the number and severity of treatment, whereas the inflammatory parameters (CRP and ESR) and disease activity 7 VF and radiographic damage increased. This observation suggests that despite responded well. A decrease in bone-resorption markers and osteoclast-regulating the decrease in inflammation and increase in the amount of bone, the anticipated markers was therefore expected alongside an increase in the bone-formation marker increase in bone quality stays behind. In addition, the ongoing bony proliferation [12, 15, 17]. Still, our results were in line with the study of Allali et al. who also 8 is also unfavourable, which emphasizes that despite TNF-α blockers bone-(patho) found in 29 AS patients an increase in BMD during treatment with TNF-α blockers physiology is still not optimal. and no change in biochemical markers (osteocalcin and total deoxypyridinoline) [31]. However, an early increase after 2-12 weeks in markers of bone formation Since persistent inflammation might be an etiologic factor of bone loss in AS, anti- (bone alkaline phosphatase) was found in other studies [15, 37], but no change in 9 TNF-α therapy has been proposed as treatment that controls inflammation with levels of CTX, OPG and RANKL [37]. Arends et al. however, showed an increase in subsequent prevention of osteoporosis and associated VF [29, 30]. This study shows BALP (bone formation marker), but also a decrease of serum collagen-telopeptide that after 2 years of etanercept, the BMD increased significantly in the lumbar spine (bone resorption marker) [17]. It is not fully clear why we did not find changes in the 10 as well as in the hips. This finding is in concordance with other studies that also investigated bone-markers. It could be related to methodological errors (samples showed an increasing trend in BMD after treatment with TNF blockers [15, 17, 31, 32]. were not taken in a fasting state), type of TNF-α blocker or type of the measured bone-markers. However until now, the result are conflicting and the value of bone Strikingly, this is the first study that describes the rapid progression of the markers is still not fully elucidated, especially in clinical practice [38]. number and severity of VF over two years, despite lowering disease activity and Interestingly, despite etanercept the radiological damage increased significantly

130 131 Chapter 9 Bone mineral density and vertebral fractures over time (p<0.001). This is confirmed by other studies with TNF blockers, including REFERENCES etanercept, which show no delay in radiological progression in AS [13, 39]. Kang et al. recently wrote about the ‘paradoxical effects’ of TNF inhibitors on BMD [1] El Maghraoui A. Osteoporosis and ankylosing spondylitis. Joint Bone Spine 2004;71:291-5. and radiographic progression in AS, since they also found an increase in BMD in [2] Karberg K, Zochling J, Sieper J, Felsenberg D, Braun J. Bone loss is detected more combination with radiographic progression of the spine, like we did [40]. Although frequently in patients with ankylosing spondylitis with syndesmophytes. J Rheumatol 1 they didn’t study VF in combination with radiological progression. Maksymowych 2005;32:1290-8. et al. hypothesized that early inflammatory lesions resolve after treatment with [3] Van der Weijden MA, Claushuis TA, Nazari T, Lems WF, Dijkmans BAC, van der Horst- Bruinsma IE. High prevalence of low bone mineral density in patients within 10 years of TNF blockers before the induction of reparative changes, whereas in more mature onset of ankylosing spondylitis: a systematic review. Clin Rheumatol 2012;31:1529-35. inflammatory lesions (visible on MRI as focal fat infiltration which reflects post 2 [4] Will R, Palmer R, Bhalla AK, Ring F, Calin A. Osteoporosis in early ankylosing spondylitis: a inflammatory tissue) new bone will be formed. New bone will be formed once the primary pathological event? Lancet 1989;2(8678-8679):1483-5. signalling pathways have been activated (through down regulation of dickkopf-1 [5] Donnelly S, Doyle DV, Denton A, Rolfe I, McCloskey EV, Spector TD. Bone mineral density which upregulates the Wnt pathway). Our study population consisted of patients and vertebral compression fracture rates in ankylosing spondylitis. Ann Rheum Dis 3 with highly active disease and a long disease duration. It could be that the resolution 1994;53:117-21. of inflammation in these more mature lesions following etanercept treatment may [6] Geusens P, Vosse D, van der Linden S. Osteoporosis and vertebral fractures in ankylosing have caused the ongoing process of new bone formation [40-42]. spondylitis. Curr Opin Rheumatol 2007;19:335-9. [7] Van der Weijden MAC, van der Horst-Bruinsma IE, van Denderen JC, Dijkmans 4 The fact that there are no studies at this moment that investigated the effects of BAC, Heymans MW, Lems WF. High frequency of vertebral fractures in early etanercept on both BMD, the occurrence of VF and radiological progression in spondylarthropathies. Osteoporos Int 2012;23:1683-90. combination with bone-markers, makes this study unique. As such this is a very [8] Gratacos J, Collado A, Filella X, Sanmarti R, Canete J, Llena J, et al. Serum cytokines clinically relevant combination of outcome measures. However, there are some (IL-6, TNF-alpha, IL-1 beta and IFN-gamma) in ankylosing spondylitis: a close correlation 5 between serum IL-6 and disease activity and severity. Br J Rheumatol 1994;33:927-31. potential limitations to be mentioned, such as the limited size of the cohort (n=49). Gratacos J, Collado A, Pons F, Osaba M, Sanmarti R, Roque M, et al. Significant loss of Nevertheless, this number is higher compared to other studies [16, 31, 40] and the [9] bone mass in patients with early, active ankylosing spondylitis: a followup study. Arthritis results are clear enough to show the challenges we are facing on this topic. Further, Rheum 1999;42:2319-24. 6 the duration of the study is limited by 2 years and the measurements of the bone- [10] Sambrook P. Tumour necrosis factor blockade and the risk of osteoporosis: back to the markers are performed within a maximum treatment duration of 1 year. However, it future. Arthritis Res Ther 2007;9:107. was to be expected that changes in biomarkers would have occurred the first year of [11] Seriolo B, Paolino S, Sulli A, Ferretti V, Cutolo M. Bone metabolism changes during etanercept treatment, since it is known to have a strong and early effect on disease anti-TNF-alpha therapy in patients with active rheumatoid arthritis. Ann N Y Acad Sci 7 activity and subsequently, also on bone-markers as has been shown in rheumatoid 2006;1069:420-7. arthritis [12]. Finally, three patients used bisphosphonates. In our opinion this has [12] Vis M, Havaardsholm EA, Haugeberg G, Uhlig T, Voskuyl AE, van de Stadt RJ, et al. not influenced the outcomes of this study, because they used it already for more Evaluation of bone mineral density, bone metabolism, osteoprotegerin and receptor 8 than 3 years and the results including these patients were not significantly different activator of the NFkappaB ligand serum levels during treatment with infliximab in patients with rheumatoid arthritis. Ann Rheum Dis 2006;65:1495-9. from when excluding them (results not shown). [13] Van der Heijde D, Landewé R, Einstein S, Ory P, Vosse D, Ni L, et al. Radiographic progression of ankylosing spondylitis after up to two years of treatment with etanercept. This prospective cohort study shows that after 2 years of TNF- blocking therapy α Arthritis Rheum 2008;58:1324-31. 9 with etanercept, BMD of the hip and spine increases, but also both the number and [14] Van der Heijde D, Landewé R, Baraliakos X, Houben H, van TA, Williamson P, et al. severity of VF. Besides that, the radiological damage, including the thoracic spine, Radiographic findings following two years of infliximab therapy in patients with increased significantly. With this study we showed that with increasing the BMD, ankylosing spondylitis. Arthritis Rheum 2008;58:3063-70. 10 bone quality is not necessarily increased. The favourable bone preserving effect is [15] Visvanathan S, van der Heijde D, Deodhar A, Wagner C, Baker DG, Han J, et al. Effects of accompanied by unfavourable outcomes on VF and radiological damage, suggesting infliximab on markers of inflammation and bone turnover and associations with bone both a lack of increase in bone strength and also a further ankylosis of the spine. More mineral density in patients with ankylosing spondylitis. Ann Rheum Dis 2009;68:175-82. attention and research is needed to investigate the aspects of quality of bone in AS [16] Marzo-Ortega H, McGonagle D, Haugeberg G, Green MJ, Stewart SP, Emery P. Bone patients and the thoracic spine may not be overlooked as important place for VF! mineral density improvement in spondyloarthropathy after treatment with etanercept. Ann Rheum Dis 2003;62:1020-1.

132 133 Chapter 9 Bone mineral density and vertebral fractures

[17] Arends S, Spoorenberg A, Houtman PM, Leijsma MK, Bos R, Kallenberg CG, et al. The [33] Lentle BC, Brown JP, Khan A, Leslie WD, Levesque J, Lyons DJ, et al. Recognizing and effect of three years of TNFalpha blocking therapy on markers of bone turnover and their reporting vertebral fractures: reducing the risk of future osteoporotic fractures. Can Assoc predictive value for treatment discontinuation in patients with ankylosing spondylitis: a Radiol J 2007;58:27-36. prospective longitudinal observational cohort study. Arthritis Res Ther 2012;14:R98. [34] Donnelly E. Methods for assessing bone quality: a review. Clin Orthop Relat Res [18] Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing 2011;469:2128-38. spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum [35] Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral 1 1984;27:361-8. density predict occurrence of osteoporotic fractures. BMJ 1996;312:1254-9. [19] Ghozlani I, Ghazi M, Nouijai A, Mounach A, Rezqi A, Achemlal L, et al. Prevalence and risk [36] Bouxsein ML. Bone quality: where do we go from here? Osteoporos Int 2003;14 Suppl factors of osteoporosis and vertebral fractures in patients with ankylosing spondylitis. 5:S118-S127. Bone 2009;44:772-6. [37] Woo JH, Lee HJ, Sung IH, Kim TH. Changes of clinical response and bone biochemical 2 [20] Longo DL, Fauci AS, Kasper DL, Hauser SL, Jameson JL, Loscalzo J, eds. Harrison’s markers in patients with ankylosing spondylitis taking etanercept. J Rheumatol principles of , 18th Edition. McGraw-Hill, New York 2012. 2007;34:1753-9. [21] Assessment of fracture risk and its application to screening for postmenopausal [38] Wheater G, Elshahaly M, Tuck SP, Datta HK, van Laar JM. The clinical utility of bone marker osteoporosis. Report of a WHO Study Group. World Health Organ Tech Rep Ser 1994;843:1- measurements in osteoporosis. J Transl Med 2013;11:201. 3 129. [39] Senabre-Gallego JM, Santos-Ramirez C, Santos-Soler G, Salas-Heredia E, Sanchez- [22] Genant HK, Wu CY, van KC, Nevitt MC. Vertebral fracture assessment using a Barrioluengo M, Barber X, et al. Long-term safety and efficacy of etanercept in the semiquantitative technique. J Bone Miner Res 1993;8:1137-48. treatment of ankylosing spondylitis. Patient Prefer Adherence 2013;7:961-72. 4 [23] Creemers MC, Franssen MJ, van’t Hof MA, Gribnau FW, van de Putte LB, van Riel PL. [40] Kang KY, Ju JH, Park SH, Kim HY. The paradoxical effects of TNF inhibitors on bone Assessment of outcome in ankylosing spondylitis: an extended radiographic scoring mineral density and radiographic progression in patients with ankylosing spondylitis. system. Ann Rheum Dis 2005;64:127-9. Rheumatology (Oxford) 2013;52:718-26. [24] Baraliakos X, Listing J, Rudwaleit M, Sieper J, Braun J. Development of a radiographic [41] Maksymowych WP. Disease modification in ankylosing spondylitis. Nat Rev Rheumatol 5 scoring tool for ankylosing spondylitis only based on bone formation: addition of the 2010;6:75-81. thoracic spine improves sensitivity to change. Arthritis Rheum 2009;61:764-71. [42] Maksymowych WP, Morency N, Conner-Spady B, Lambert RG. Suppression of [25] Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to inflammation and effects on new bone formation in ankylosing spondylitis: evidence for a defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease window of opportunity in disease modification. Ann Rheum Dis 2013;72:23-8. 6 Activity Index. J Rheumatol 1994;21:2286-91. [26] Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mallorie P, et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index. J Rheumatol 1994;21:2281-5. 7 [27] Jenkinson TR, Mallorie PA, Whitelock HC, Kennedy LG, Garrett SL, Calin A. Defining spinal mobility in ankylosing spondylitis (AS). The Bath AS Metrology Index. J Rheumatol 1994;21:1694-8. [28] Braun J, Pham T, Sieper J, Davis J, van der Linden S, Dougados M, et al. International 8 ASAS consensus statement for the use of anti-tumour necrosis factor agents in patients with ankylosing spondylitis. Ann Rheum Dis 2003;62:817-24. [29] Kawai VK, Stein CM, Perrien DS, Griffin MR. Effects of anti-tumor necrosis factor alpha agents on bone. Curr Opin Rheumatol 2012;24:576-85. 9 [30] Maillefert JF, Aho LS, El Maghraoui A, Dougados M, Roux C. Changes in bone density in patients with ankylosing spondylitis: a two-year follow-up study. Osteoporos Int 2001;12:605-9. 10 [31] Allali F, Breban M, Porcher R, Maillefert JF, Dougados M, Roux C. Increase in bone mineral density of patients with spondyloarthropathy treated with anti-tumour necrosis factor alpha. Ann Rheum Dis 2003;62:347-9. [32] Kang KY, Lee KY, Kwok SK, Ju JH, Park KS, Hong YS, et al. The change of bone mineral density according to treatment agents in patients with ankylosing spondylitis. Joint Bone Spine 2011;78:188-93.

134 135 CHAPTER 10

1 General discussion and summary 2 Nederlandse samenvatting 3 Dankwoord 4 Curriculum vitae 5 List of publications 6 List of abbreviations and illustrations Chapter 10 General discussion and summary

1. GENERAL DISCUSSION AND SUMMARY symptomatic relief but also have anti-inflammatory properties is supported by the decrease of acute phase reactants (erythrocyte sedimentation rate (ESR)) and a Evolving treatment of Ankylosing Spondylitis. small effect on axial MRI bone lesions during treatment with in an open- The therapy of ankylosing spondylitis (AS) consisted for a long time of patient label study [3]. Continuous use of a NSAID, in contrast to on-demand use, proofed education, exercise therapy and non-steroidal anti-inflammatory drugs (NSAIDs) to retard radiographic ossification of the lumbar spine, especially in patients with 1 and nowadays these are still the cornerstones of treatment. elevated acute phase reactants [4-6]. Hence, in addition to analgetic and antiflogistic Traditionally, NSAIDs are anti-inflammatory drugs that mainly affect symptoms. effects, NSAIDs might have disease modifying properties. Inhibition of bone In contrary, the role of disease modifying antirheumatic drugs (DMARDs), which formation was seen also in the prevention of heterotopic ossification by coxibs and are characterized by changing the course of the disease and delaying structural effect on bone healing [4, 7]. Apparently paradoxical, NSAIDs seem to improve bone 2 damage, was always limited in AS. Over ten years now, the tumour necrosis mineral density [8]. Considering the continuous use of a NSAID, the likelihood of factor (TNF)-blockers have dramatically changed the therapeutic options in AS and radiographic progression in an individual patient has to be taken into account and spondyloarthritis (SpA). weighted against potential (gastrointestinal and cardiovascular) adverse events [9]. The best predictor of structural progression is the presence of syndesmophytes at 3 However, the traditional distinction between NSAIDs and DMARDs is not always baseline [10]. In contrast, high levels of functional dickkopf-1 predicts protection clear in AS. There is some evidence that NSAIDs do not only modify symptoms, from syndesmophyte formation [11]. but also seem to delay radiographic progression. On the other hand the effect of 4 DMARDs was often primarily measured as effect on symptoms and inflammation In summary NSAIDs are still the cornerstone of the treatment of AS. Continuous use and it was doubted if TNF-blockers would have effect on radiographic progression. In is recommended in patients with elevated acute phase reactants and progressive addition, the clinical relevance of the growth of some syndesmophytes is debatable disease. in patients who experience impressive relief of symptoms, following TNF-blocking 5 therapy [1]. However, for clarity, in this chapter the traditional denomination of DMARDs: sulfasalazine and mesalazine. NSAIDs and DMARDs, as used in rheumatoid arthritis (RA), was followed. Sulfasalazine is not effective in AS with only spinal symptoms, but several studies showed some efficacy of sulfasalazine in active AS/SpA with peripheral arthritis 6 The first part of this thesis consists of studies performed just before the TNF-blockers (weighted mean difference in ESR -4.79 and morning stiffness VAS-100mm -13.89 became available and the second part concerns several aspects of anti-TNF therapy compared to placebo) [12-15]. Presently, sulfasalazine is still the initial treatment itself. The TNF-blockers seem to have solved many of the problems in AS patients, option for AS patients with peripheral involvement [2, 16]. but still many questions about drug therapy in AS remain outstanding: 7 Do NSAIDs really have disease modifying effects? What is the role of DMARDs in Sulfasalazine is split in the large intestine in sulfapyridine and mesalazine AS? How do we assess the efficacy of medication? What is the effect of medication (5-aminosalicylic acid), having antibacterial and antiflogistic properties, respectively. on extra-articular manifestations, especially anterior uveitis, and what is the effect Mesalazine is the active drug in the treatment of inflammatory bowel disease. 8 on bone and cardiovascular disease? What is the effect of TNF-blockers and what are The effect of mesalazine was investigated in 20 patients with active AS in an open the main adverse events? What can we say about non-responders and do we know study for 24 weeks (chapter 2). There was a high rate of premature discontinuation enough about long-term therapy? Is there still an unmet need for new drugs to treat because of intolerance. Except for a small significant decrease in erythrocyte AS? sedimentation rate, treatment with mesalazine did not show improvement in other 9 These various aspects of drug therapy in AS are discussed in this chapter. disease variables. Summaries of the previous chapters of this thesis and conclusions are shown in cursive script. The mode of action of sulfasalazine in AS is unknown. However in 1985, a high rate 10 of incidence of inflammation in the intestine was found in patients with HLA-B27 NSAIDs. related arthritis [17]. In more recent studies, these numbers of intestinal involvement Non-steroidal anti-inflammatory drugs (NSAIDs), including cyclo-oxygenase-2- were confirmed in axial SpA patients [18, 19], which might explain the beneficial selective inhibitors (coxibs), are recommended as first-line therapy for inflammatory effects of sulfasalazine. Moreover the bacterial flora of the gut seems to be an back pain and stiffness in patients with AS [2]. Evidence that NSAIDs not only give important factor in the role of HLA-B27 in the pathogenesis of AS [20]. When the role

138 139 Chapter 10 General discussion and summary

of the bowel and its flora is more elucidated in AS, the intestines and the enteral These data were confirmed by another small and open label study. In this study microbes could again be important as targets for treatment. axial symptoms did not improve, but in patients suffering from peripheral arthritis the number of inflamed joints was reduced with leflunomide treatment [31]. The In AS patients with peripheral arthritis sulfasalazine is recommended. There is no number of patients with peripheral arthritis in our trial was too small to draw evidence for use of only mesalazine in these cases. any conclusion about the efficacy of leflunomide in this subgroup. If a difference 1 in efficacy between these subgroups would exist, this might be explained by Statins as antirheumatic drugs. differences in pathophysiology. Significant higher MMP-3 concentrations, for Statins have anti-inflammatory properties, as demonstrated by lowering C-reactive example, are found in AS patients with peripheral disease compared to those with protein and disease activity in rheumatoid arthritis [21]. only axial disease [32]. 2

The effect of the statin rosuvastatin in AS was investigated in 15 consecutive In general, there is no place for leflunomide in the treatment of AS, but leflunomide patients with active disease in an open study during 12 weeks, followed by an might be considered in patients with peripheral arthritis resistant to other therapies. observational period of 12 weeks (chapter 3). Treatment with rosuvastatin resulted 3 in significant decreases of C-reactive protein and erythrocyte sedimentation Other DMARDs. rate after 12 weeks and a trend of improvement in several clinical variables. A Corticosteroids are effective when used intra-articular in peripheral or sacroiliac randomised, placebo controlled trial is required to prove the anti-inflammatory and joints [33]. Oral corticosteroids were seldom studied in AS. A recent study showed 4 disease modifying properties of statins in AS. As could be expected, total and LDL- short-term efficacy of a high dose prednisone when given for two weeks [34]. A 50% cholesterol levels were significantly reduced during treatment in this study. improvement of the Bath AS Disease Activity Index (BASDAI) was seen in 33% of the patients treated with 50 mg prednisone every day, versus 8% on placebo. More The improvement of the lipid profile is a favourable effect of statin therapy in AS, studies are needed to establish the role of corticosteroids. An option could be that 5 particularly in view of the increased cardiovascular risk in AS patients [22, 23]. corticosteroids are used in AS as bridging therapy, like in RA, awaiting the efficacy of Concerning this increased cardiovascular risk in AS patients, attention is also needed another treatment. for cardio-respiratory fitness and other traditional risk factors for cardiovascular 6 disease, as smoking, high blood pressure, diabetes and overweight [24, 25]. Methotrexate (MTX) was studied in a few randomized controlled AS trials, but in low Moreover, effective treatment of inflammation has favourable effects on the numbers of patients and only in low doses (7.5-10 mg/week). Thus far, insufficient cardiovascular risk in AS [26-28]. evidence was found to support benefit of MTX in the treatment of AS [35]. Studies with higher doses of MTX are needed, also in the light of the possible role of MTX in 7 In addition to treatment of inflammation, in AS patients attention is recommended reducing immunogenicity, as is discussed below. for traditional risk factors for cardiovascular disease. Further studies on antirheumatic properties of statins are needed. A short course of therapy with corticosteroids could be considered in AS, but more 8 studies are needed to investigate long-term treatment with corticosteroids and high DMARDs: leflunomide. doses of methotrexate. Leflunomide is proven to be effective in rheumatoid arthritis and psoriatic arthritis, but studies in AS were lacking so far [29, 30]. TNF-blockers 9 From the beginning of this century tumour necrosis factor alpha (TNF) blockers were The effect of leflunomide in AS was studied in a double blind, randomised, placebo used in AS and proofed to be very effective on both a short-term and long-term controlled trial in 45 patients with active disease (chapter 4). The number of basis. During treatment with TNF-blockers, improvement was seen in symptoms, 10 responders according to the Assessment of SpondyloArthritis international Society including spinal pain, morning stiffness, functioning, spinal mobility, arthritis, (ASAS) 20% definition was higher in the leflunomide group than in the placebo enthesitis, extra-articular manifestations, sleep, quality of life and also in acute group, but the difference did not reach statistical significance. Although the study phase response, hemoglobin levels and inflammatory MRI lesions [1, 36-38]. was underpowered, the differences between the treatment groups were small and Currently, five TNF-blockers are available for treatment of AS: infliximab, etanercept, therefore striking differences in a trial with higher numbers of patients are not likely. adalimumab, golimumab and certolizumab [39-43]. In a meta-analysis the odds ratio

140 141 Chapter 10 General discussion and summary

for being in ASAS40 response was 4.7 (95% CI 3.8 - 6.0) for patients taking a TNF- Serum drug levels and antidrug antibodies were studied in 38 AS patients treated blocker compared with placebo [44]. with infliximab for 54 weeks chapter( 5). At the end 53% of the patients met the 20% ASAS response criteria. The mean serum trough infliximab level for responders was The ASAS has published international recommendations for the use of TNF- significantly higher compared to the non-responders (8.2 mg/l versus 6.3 mg/l). In blockers in patients with axial spondyloarthritis [45]. Treatment with TNF-blockers addition, anti-infliximab was significantly more often found in non-responders (59% 1 is considered the standard of care for AS patients with active disease, who fail to versus 5%). As shown in other studies, antibodies against the TNF-blocker were in respond adequately to conventional treatment. Active disease is defined as at least our study associated with increased risk of infusion reactions [51]. four weeks active disease, according to a Bath AS Disease Activity Index (BASDAI) score of at least 4 (0-10) and positive expert opinion. Conventional therapy is defined Antidrug antibodies, associated with reduced treatment response, were also 2 as treatment with at least two NSAIDs over a 4-week period in total, at maximum shown in 31% of 35 AS patients treated with adalimumab for six months [52]. In recommended or tolerated anti-inflammatory dose unless contraindicated. RA, co-administration of methotrexate (MTX) reduces the immunogenicity of the Moreover, patients with symptomatic peripheral arthritis must have had at least TNF-blocker in a dose-dependent manner. The overall Odds Ratio was 0.20 (95% one local steroid injection and should normally have had a therapeutic trial of a CI 0.12-0.34) to develop antibodies against adalimumab in RA patients using MTX 3 DMARD, preferably sulfasalazine. Patients with peripheral enthesitis must have compared with patients without MTX [53]. In AS, the same effect was suggested in failed appropriate local treatment. The effect of treatment with a TNF-blocker is a small open label controlled study in which infliximab in combination with MTX evaluated after at least 12 weeks. The response criteria are a 50% relative change or seemed to be more efficacious than infliximab alone [54]. 4 absolute change of 2 (0-10 scale) in BASDAI score and expert opinion in favour of Despite the fact that evidence for effect of MTX on AS disease activity is lacking, continuation. it might be interesting to conduct studies on the effect of MTX and other immunosuppressive drugs on antibody formation against TNF-blockers and the In a questionnaire among AS patients, recruited from a secondary care clinic, 64% influence on the response rate in AS. 5 reported a BASDAI score of at least 4 (scale 0-10). The authors concluded that there is a large unmet need for TNF-blocking therapy in AS, although an expert opinion In case of insufficient response or adverse events during TNF-blocking treatment, was not available in most of these patients [46]. In a representative group of Belgian switching to another TNF-blocker can be effective. Overall the response rates among 6 rheumatology offices about 40% of the AS patients were supposed to be eligible for switchers are lower, but about half of them achieve treatment response [55-57]. anti-TNF therapy, for the most part according to the international recommendations Information about drug serum levels or antidrug antibodies could be helpful in these [47]. situations. A main disadvantage of this successful treatment are the high costs. In 2011 two 7 TNF-blockers were on the first and second place of the top ten of expenses for drugs Antidrug antibodies, associated with reduced treatment response, are found in delivered by the public in the Netherlands [48]. The expenses for the 29-31% of AS patients treated with infliximab or adalimumab. Further studies are two drugs together were more than 350 million euros (total drug expenses 5001 needed to determine the effect of immunosuppressive drugs on antibody formation 8 million euros). Next to the registered TNF-blockers, so-called biosimilars have been and to define the role of antibody determination in clinical decision making. developed, which may be more cost-effective. The first studies with a biosimilar of infliximab in AS showed equivalent efficacy compared to the original TNF-blocker Adverse events of TNF-blockers. [49]. In placebo controlled trials, the most frequent adverse events of TNF-blockers 9 were injection site reactions (or infusion reactions) and mild infections [39-43]. In TNF-blockers are often effective for most of the signs and symptoms of AS and have this paragraph the risks of serious infections, malignancy and liver problems are dramatically broadened the therapeutic arsenal in AS. discussed in more detail. 10

Immunogenicity of TNF-blockers. In assessing the risk of serious infections one has to keep in mind that patients with In RA it has been shown that several TNF-blockers can induce an immune response immune mediated inflammatory disorders have an increased risk of infections not and the formation of antidrug antibodies. The antibodies were associated with low only due to medication, but also because of the disease itself. In a meta-analysis, RA serum drug levels and reduced response to treatment [50]. patients on TNF-blockers had an increased risk of serious infections (adjusted hazard

142 143 Chapter 10 General discussion and summary

ratios (aHR) between 1.1 and 1.8) and tuberculosis (aHR 12.5) versus patients on vulnerability of a liver with steatosis, whereas on the other hand a favourable effect conventional DMARDs [58]. of TNF-blockers on fatty liver disease is expected [67]. In AS, the incidence rate of serious infections was found to be lower compared to RA and Crohn’s disease [59]. Perhaps this is related to the lower use of The definite risk of hepatic disturbance during treatment of TNF-blockers can only immunosuppressive drugs in AS. In a systematic literature review in AS, the be determined by comparison with a control group. In some placebo controlled 1 absolute risk of serious infections in patients not exposed to TNF-blockers was low. trials of several TNF-blockers, increases of liver enzymes were seen 2-3 times more The risk of serious infections in patients receiving TNF- blockers was higher, but the frequent in patients using the TNF-blocker than those using placebo [41, 42, 68]. Also difference was not significant, possibly due to a lack of power [60]. the mean increase of liver enzymes was higher compared to the placebo-treated patients and therefore it is likely to be ascribed to TNF-blockers. 2 Next to the role of TNF in host defence, it plays an important role in the pathobiology of cancer [61]. Therefore, malignancy has been considered as a possible adverse The most important adverse event of TNF-blockers is an increased risk of infections. event of TNF-blockers. Regarding the risk of malignancy during TNF-blocking therapy, This risk seems to be lower in AS patients than in RA. To assess the risk of the possibility of a higher risk for certain malignancies by the inflammatory disease malignancy in AS patients, treated with a TNF-blocker, more studies in large cohorts 3 itself has to be taken into account. Patients with RA appear to be at higher risk of are required. lymphoma and cancer compared with the general population, but overall the Increase of liver enzymes in TNF-blocker treated AS patients was seen in 9% malignancy risk is almost similar [62]. of the patients. Therefore, regular testing of liver enzymes in these patients is 4 Data on a possible increased risk of malignancy during treatment with TNF-blockers recommended (for example every three months in the first year of therapy). Patients are conflicting, but in a Swedish register TNF-blockers were not associated with a with higher BMI appear to be more prone to this side effect. further increase of the risk for lymphoma in RA [63]. In another study TNF-blocking therapy in RA was associated with an increased risk for non-melanoma skin cancer Measuring physical function and TNF-blocking. 5 [64]. The efficacy of TNF-blockers in AS is determined by the improvement in several Less is known about the incidence of malignancy in AS patients, with and without disease parameters during treatment. The most important outcome parameters TNF-blocking therapy. In a Belgian cohort, a tendency towards a higher incidence in AS are disease activity, function, spinal mobility and quality of life [1]. These 6 of malignancy in SpA patients treated with a TNF-blocker was seen, compared to parameters are interrelated and are assessed for a great part by self-reported, the general population. However, the numbers were too small to draw definite disease-specific, visual analogue scale (VAS) based questionnaires and physical conclusions [61]. In a Taiwanese cohort with matched controls, an increased risk functioning is commonly assessed with the Bath AS Functional Index (BASFI) of cancer, particularly lung or head and neck cancer, (aHR 1.38) was found in TNF- questionnaire [69, 70]. However self-reported outcome measures are susceptible to 7 blocker naïve AS patients [65]. The risk of malignant lymphoma in AS was not under- or overestimation [71]. elevated in a national register and not affected by the use of TNF-blockers [66]. The overall malignancy rates for adalimumab-treated patients with AS or other Previous studies suggested that performance-based tests of functioning and the 8 inflammatory disorders was as expected for the general population in a meta- BASFI questionnaire are only moderately associated, indicating that these methods analysis of clinical trials [59]. measure different aspects of physical functioning [72]. Further evaluating this subject, the improvement of performance-based assessment of physical function Serious liver problems were rarely seen in trials with TNF-blockers [40, 43]. [Gorman, was established in 82 patients treated with a TNF-blocker for three months (chapter 9 2002; Landewé, 2014] In clinical practice, however we observed several AS patient 7). Improvement of performance-based physical function tests was seen in 48% with liver enzyme elevations during treatment with etanercept. This prompted us of the patients that were categorized as non-responders according to the ASAS20 to study this systematically (chapter 6). Of 105 consecutive AS patients treated with response criteria and in 56% of the patients showing no improvement in self- 10 etanercept for at least three months, 15 patients had significant elevation of liver reported physical functioning. Therefore the discrepancy between a performance- enzymes (defined as more than 1.5 times the upper normal limit) more than once. In based and a questionnaire-based assessment of response was confirmed. nine cases the elevation was probably or possibly related to treatment with the TNF- blocker. An increased risk of elevation of liver enzymes was found in patients with a Additional studies are needed to show the clinical relevance and the underlying higher body mass index (BMI). This observation might be explained by an increased mechanisms of these differences. In knee-osteoarthritis it was observed that self-

144 145 Chapter 10 General discussion and summary

reported physical function was more influenced by pain than by performance- soluble receptor blocker with respect to the risk of tuberculosis during therapy with based function [73]. In AS, the influence of cardiovascular fitness on perceived TNF-blockers [82]. disease activity was seen and, in AS patients with peripheral joint impairment, Despite the positive effect of TNF-blockers in general on extra-articular two questionnaires of functioning performed differently [74, 75]. Another study manifestations, an apparent paradoxical effect can sometimes be seen when uveitis, showed worse BASFI scores in AS patients compared with control subjects, but they psoriasis or inflammatory bowel disease occur during treatment with TNF-blockers 1 reported and objectively performed the same amount of physical activity [76]. [83-85]. It is proposed to use a combination of measurements in evaluating different aspects of functioning of AS patient, keeping in mind feasibility. Similarly, a lack of TNF-blockers have favorable effects on several extra-articular manifestations of AS, correlation is sometimes seen between spinal inflammation and reported level of but in this respect the several TNF-blockers differ in efficacy. 2 disease activity [77]. Therefore, disease activity in AS is preferably assessed by the Ankylosing Spondylitis Disease Activity Score (ASDAS), which includes an objective Bone and TNF-blockers. measure of inflammation (ESR or CRP) in contrast to the BASDAI questionnaire [78]. Interestingly, AS is characterized by the paradox of new bone formation (syndesmophytes and ankylosis) on the one hand and bone loss and increased risk 3 Disease parameters in AS show correlations as well as discrepancies between self- of fractures on the other hand [8]. reported and performance-based tests. Performance-based tests should be further developed. For good understanding of the cause and degree of the problems of the The influence of a TNF-blocker on bone was studied in 49 AS patients treated with 4 patient several tests have to be used, including performance-based tests. etanercept (chapter 9). After two years of treatment the bone mineral density (BMD) of hip and spine was increased significantly (2.2%, resp. 7.0%). However, an increase Extra-articular manifestations and TNF-blockers. was also seen in vertebral fractures (from 6 to 15 patients) and new bone formation In AS extra-spinal and extra-articular manifestations regularly occur, in particular (radiological score from 12.1 to 18.5). No significant changes were found in several 5 psoriasis, inflammatory bowel disease and anterior uveitis. One or more extra- markers of bone turnover. articular manifestations were seen in 42% of AS patients in a Belgian study [47]. Other studies found even higher incidences and extra-articular manifestations can In AS a decreased BMD is seen, although measurement of BMD of the spine 6 also affect the bone, heart, lung and [79]. with dual-energy X-ray absorptiometry (DEXA), in contrast to the hip, can be The mean prevalence of anterior uveitis was 33% in a systemic review [80]. The overestimated in case of new bone formation along the spine. A decreased BMD anterior uveitis in AS is usually acute, unilateral and often recurrent. Inadequately in AS is probably related to inflammatory factors. The expected positive effect of treated the sequels for the eye can be vision-threatening. The primary therapy of strong anti-inflammatory drugs, like TNF-blockers, on BMD was also demonstrated 7 anterior uveitis is treatment by the ophthalmologist with topical corticosteroids in in other studies [86, 87]. It is unknown what the influence of TNF-blockers is on the combination with pupil dilating eye drops. ultimate outcome, namely fractures (without or with symptoms). In our study the number of vertebral fractures increased during treatment with the TNF-blocker, but a 8 The effect of the TNF-blocker adalimumab on the occurrence of anterior uveitis was control group was lacking. studied in 77 consecutive AS patients with active axial disease treated for at least In addition, there is lack of knowledge about the effect of anti-osteoporotic drugs 12 weeks (chapter 8). Compared to the year before treatment, the number of attacks like bisphophonates on BMD and fractures in AS. The effect of bisphosfonates is also per 100 patient years was 80% reduced and the number of uveitis attacks per year interesting because of the claimed anti-inflammatory activity in AS [88-90]. 9 decreased with 72%. TNF-blockers reduce MRI-detected spinal inflammation [38]. Previously, it was TNF-blockers appear to differ in the degree of their effect on extra-articular demonstrated that inflammatory MRI lesions predict the development of new 10 manifestations, especially uveitis and inflammatory bowel disease [81]. Part of syndesmophytes, although other studies show a less clear association [91, 92]. this difference could be explained by the observation that the soluble receptor Despite the anti-inflammatory effects of TNF-blockers, new bone formation during blocker etanercept has less effect on granulomas than the other TNF-blockers, which treatment was seen in comparison with the progression in TNF-blocker naïve are monoclonal antibodies. This might explain why etanercept is not effective in historical cohorts [93-95]. These somewhat disappointing results raise questions inflammatory bowel disease. On the other hand this difference is favourable for the about the mechanism of (un)coupling of inflammation and bone formation [87]. It

146 147 Chapter 10 General discussion and summary

is hypothesized that, although TNF-blockers can decrease inflammation, they do not Duration of treatment with TNF-blockers? directly inhibit the formation of new bone. Bone formation might be the result of a Several studies show persistent favourable clinical results in AS patients treated repair mechanism, driven by mesenchymal tissue responses which are triggered by with TNF-blockers for many years [111, 112]. Early remission is the best predictor for the resolution of inflammation and perhaps also triggered by mechanical stress [96]. long-term remission through 5 years of treatment [113]. Recent long-term studies however, demonstrated that finally less pathological bone What is known about the necessary duration of therapy with a TNF-blocker? 1 formation occurs in AS patients treated with TNF-blockers, probably because of the Discontinuation of the TNF-blocker in a patient with a favourable response seems reduction of new inflammatory lesions [97-99]. to lead to a relapse in many cases [114, 115]. However, in part of the patients (32- More studies are needed to unravel the effect of early treatment on radiographic 47 %) it is presumably possible to maintain low disease activity with TNF-blockers progression before irreversible changes have occurred. Moreover, it would be in reduced doses [116, 117]. Dose reduction might be possible in case of high 2 interesting to study the effect of continuous use of NSAIDs or other therapies on the serum drug levels. Predictive characteristics for successful discontinuation or dose progression of new bone formation during (the first years of) treatment with TNF- reduction are not yet known; disease duration, age or duration of remission could be blockers. This should be done by controlled and long-term studies, because of the such factors. slow rate of radiographic progression in AS [100]. 3 Another question is whether non-radiographic axial SpA patients are less prone During treatment with TNF-blockers bone mineral density increases in AS, but so far to develop a flare after stopping TNF-blocking therapy than patients with definite there is no evidence that TNF-blockers reduce the number of new vertebral fractures. AS. After two years follow-up only 8% of patients with early axial SpA reached 4 Radiographic progression in AS is probably retarded by TNF-blockers on the long- permanent drug-free remission after treatment with etanercept for one year [118]. term. In 24 patients with non-radiographic axial SpA and good response after treatment with adalimumab for 1 year, only 17% experienced no disease flare during 2 years Early treatment with TNF-blockers? after withdrawal of the drug [119]. However, in another study about 43% of the axial 5 TNF-blocking has shown to be effective in patients with early axial SpA in SpA patients remained in remission during 6 months in those patients who reached comparison with placebo or NSAID, although in many of these studies a partial remission after treatment for 28 weeks with either infliximab plus naproxen considerable amount of the patients already meet the radiographic criteria of AS or naproxen alone in the preceding period [120]. 6 [43, 101-104]. In case of high disease activity, AS can be treated with TNF-blockers An attempt to discontinue treatment in good responders seems justified, because in the early (pre-radiographic) phase of the disease and possibly radiographic the response to retreatment with the TNF-blocker seems well in AS and non- progression can best be retarded by early treatment. The classification criteria of radiographic axial SpA [119, 121]. axial spondyloarthritis (SpA) can be helpful to diagnose and treat the disease early 7 [45]. Another argument to treat non-radiographic axial SpA and AS in the same way, Relapse of the disease is seen in many AS/SpA patients after discontinuation of is their similar burden of disease, defined as disease activity, functional impairment successful therapy with a TNF-blocker. However, dose reduction and discontinuation and quality of life [105, 106]. should be attempted and studied further. 8 On the other hand little is known about the long term prognosis of the whole group of axial SpA [107]. Not all cases of non-radiographic axial SpA finally evolve Other biologicals? to definite AS and partially non-radiographic axial SpA has features of a distinct Overall, about 30% of the AS patients do not respond to treatment with TNF- disease entity [108, 109]. When considering therapy in axial SpA patients, it has to blockers. This might be explained in several ways. One reason could be that 9 be taken in account that a higher response rate is seen in patients with radiographic the patient has no active inflammation and the symptoms are caused by other features and in patients with objective signs of inflammation, reflected in MRI mechanisms. This is not always clear, because objective signs of inflammation are lesions or acute phase response [102, 104, 106, 110]. often absent in AS. In many patients blood inflammatory markers are low and MRI 10 is not always conclusive. Another explanation might be that the inflammation is Patients with (early) non-radiographic axial SpA can be treated with TNF-blockers, TNF-driven, but the TNF-blocker is not tolerated or TNF is not adequately blocked, especially when objective signs of inflammation are present as MRI lesions or raised for example because of antibodies against the drug. Also, the role of TNF in the acute phase response. inflammation could be limited. Therefore, there is need for disease modifying drugs which modulate other pathways of inflammation than TNF-blockade.

148 149 Chapter 10 General discussion and summary

Open label trials failed to demonstrate efficacy of anakinra (IL-1 antagonist) and TNF-blockers are an important advancement in the treatment of AS. Accurate abatacept (T cell inhibition) in AS [122, 123]. The B cell inhibitor was not (nationwide) registration of efficacy, extra-articular manifestations and adverse effective in an open trial, although effect was seen in the subgroup of TNF-blocker events of the different TNF-blockers and biosimilars would be very informative. naïve patients [124]. Placebo-controlled trials failed to show efficacy of The role of antibodies against TNF-blockers has to be elucidated and especially their in AS [125]. Another interleukin (IL)-6 inhibitor, , neither showed role in predicting successful dose reduction, stopping and switching of therapy. 1 efficacy in a placebo-controlled study in AS [126]. , an oral inhibitor Performance-based tests should be further developed and incorporated in the phosphodiesterase 4, showed some improvement compared to placebo in a pilot current outcome parameters of efficacy of therapies. study, but further testing is needed [127]. Long-term follow-up is needed of patients with axial SpA, especially those without Ustekinumab, a monoclonal antibody that inhibits IL-12 and IL-23, proofed to be radiographic signs, because the long term outcome of non-radiographic SpA is 2 efficacious in treating psoriatic arthritis [128]. In AS ustekinumab reduced signs and largely unknown. symptoms in an open-label study [129]. Presently, the most promising results in It would be very interesting to investigate the characteristics of patients not AS patients are found in blocking of IL-17. In animal models intestinal interleukin-23 responding to treatment with a TNF-blocker. For part of these non-responders and subsequent stimulation of IL-17/22 was found to be involved in development of studies for other (biological) therapies are needed. Undoubtedly ongoing 3 spondyloarthritis syndrome [130]. TNF-independent increase in IL-17-expressing mast pathophysiologic research is needed in the search for new therapies. cells may contribute to synovial inflammation in peripheral SpA [131]. The anti-IL-17a antibody secukinumab was superior to placebo in reducing signs of AS in a small, 4 randomized, double-blind study [132].

Because not all AS patients respond to therapy with a TNF-blocker, there is still need for other therapies. Most promising are results of agents blocking interleukin 5 23/17, but larger controlled studies are needed for further evaluation of these new therapeutic options. 6 Some future research suggestions. NSAIDs are still the cornerstone of the drug treatment of AS. Prospective long-term studies are needed comparing continuous use with on demand use. The rate of radiographic progression in these groups can be compared, as well as the adverse 7 events. This can be done in patients without TNF-blocking therapy, as well as in patients using TNF-blockers, because after the start of TNF-blocking treatment new bone formation is still going on. 8

Sulfasalazine is used in AS patients with peripheral arthritis. Furthermore, the role of the traditional DMARDs in AS is limited. There is an unmet need for controlled studies with higher doses of methotrexate, eventually in combination with a pulse 9 dose of corticosteroids, to determine the effect on disease activity and also the effect on antibody formation against TNF-blockers. The antirheumatic properties of statins have to be studied in a controlled trial, as 10 well as the effect of cardiovascular risk screening and management in AS. The effects of bisphosphonates in AS on bone mineral density, as well as on disease activity and radiographic progression have to be studied in placebo controlled trials.

150 151 Chapter 10 General discussion and summary

REFERENCES CHAPTER 10.1

[1] Braun J, Sieper J. What is the most important outcome parameter in ankylosing [16] Fagerli KM, van der Heijde D, Heiberg MS, Wierød A, Kalstad S, Rødevand E, et al. Is spondylitis? Rheumatology 2008;47:1738-40. there a role for sulphasalazine in axial spondyloarthritis in the era of TNF inhibition? [2] Braun J, van den Berg R, Baraliakos X, Boehm H, Burgos-Vargas R, Collantes-Estevez Data from the NOR-DMARD longitudinal observational study. Rheumatolgy(Oxford) E, et al. 2010 update of the ASAS/EULAR recommendations for the management of 2014;53:1087-94. 1 ankylosing spondylitis. Ann Rheum Dis 2011;70:896-904. [17] Mielants H, Veys EM, Cuvelier C, De Vos M, Botelberghe L. HLA-B27 related arthritis [3] Jarrett SJ, Sivera F, Cawkwell LS, Marzo-Ortega H, McGonagle D, Hensor E, et al. MRI and and bowel inflammation. Part 2. Ileocolonoscopy and bowel histology in patients with clinical findings in patients with ankylosing spondylitis eligible for anti-tumour necrosis HLA-B27 related arthritis. J Rheumatol 1985;12:294-8. factor therapy after a short course of etoricoxib. Ann Rheum Dis 2009;68:1466-9. [18] Van Praet L, Van den Bosch FE, Jacques P, Carron P, Jans L, Colman R, et al. Microscopic 2 [4] Wanders A, van der Heijde D, Landewé R, Béhier JM, Calin A, Olivieri I, et al. gut inflammation in axial spondyloarthritis: a multiparametric predictive model. Ann Nonsteroidal antiinflammatory drugs reduce radiographic progression in patients with Rheum Dis 2013;72:414-7. ankylosing spondylitis: a randomized clinical trial. Arthritis Rheum 2005;52:1756-65. [19] Van Praet L, Jans L, Carron P, Jacques P, Glorieus E, Colman R, et al. Degree of bone [5] Poddubnyy D, Rudwaleit M, Haibel H, Listing J, Märker-Hermann E, Zeidler H, et al. Effect marrow oedema in sacroiliac joints of patients with axial spondyloarthritis is linked 3 of non-steroidal anti-inflammatory drugs on radiographic spinal progression in patients to gut inflammation and male sex: results from the GIANT cohort. Ann Rheum Dis with axial spondyloarthritis: results from the German Spondyloarthritis Inception Cohort. 2014;73:1186-9. Ann Rheum Dis 2012;71:1616-22. [20] Rosenbaum JT, Davey MP. Time for a gut check: evidence for the hypothesis that HLA-B27 4 [6] Kroon F, Landewé R, Dougados M, van der Heijde D. Continuous NSAID use reverts predisposes to ankylosing spondylitis by altering the microbiome. Arthritis Rheum the effects of inflammation on radiographic progression in patients with ankylosing 2011;63:3195-8. spondylitis. Ann Rheum Dis 2012;71:1623-9. [21] McCarey DW, McInnes IB, Madhok R, Hampson R, Scherbakov O, Ford I, et al. Trial [7] Fransen M, Neal B. Non-steroidal anti-inflammatory drugs for preventing heterotopic of Atorvastatin in Rheumatoid Arthritis (TARA): double-blind, randomised placebo- 5 bone formation after hip arthroplasty. Cochrane Database Syst Rev 2004;3:CD001160. controlled trial. Lancet 2004;363:2015-21. [8] Vosse D, Landewé R, van der Heijde D, van der Linden S, van Staa T-P, Geusens P. [22] Peters MJ, van der Horst-Bruinsma IE, Dijkmans BAC, Nurmohamed MT. Cardiovascular Ankylosing spondylitis and the risk of fracture: results from a large primary care-based risk profile of patients with spondylarthropathies, particularly ankylosing spondylitis and nested case-control study. Ann Rheum Dis 2009;68:1839-42. psoriatic arthritis. Semin Arthritis Rheum 2004;34:585-92. 6 [9] Haroon N, Kim TH, Inman RD. NSAIDs and radiographic progression in ankylosing [23] Chou CH, Lin MC, Peng CL, Wu YC, Sung FC, Kao CH, et al. A nationwide population- spondylitis: Bagging big game with small arms? Ann Rheum Dis 2012;71:1593-5. based retrospective cohort study: increased risk of acute coronary syndrome in patients with ankylosing spondylitis. Scand J Rheumatol 2014;43:132-6. [10] Van Tubergen A, Ramiro S, van der Heijde D, Dougados M, Mielants H, Landewé R. Development of new syndesmophytes and bridges in ankylosing spondylitis and their [24] Peters MJ, Symmons DP, McCarey D, Dijkmans BAC, Nicola P, Kvien TK, et al. EULAR 7 predictors: a longitudinal study. Ann Rheum Dis 2012;71:518-23. evidence-based recommendations for cardiovascular risk management in patients with rheumatoid arthritis and other forms of . Ann Rheum Dis [11] Heiland GR, Appel H, Poddubnyy D, Zwerina J, Hueber A, Haibel H. High level of 2010;69:325-31. functional dickkopf-1 predicts protection from syndesmophyte formation in patients with ankylosing spondylitis. Ann Rheum Dis 2012;71:572-4. [25] Halvorsen S, Vøllestad NK, Provan SA, Semb AG, van der Heijde D, Hagen KB, et al. 8 Cardiorespiratory fitness and cardiovascular risk in patients with ankylosing spondylitis: [12] Dougados M, van der Linden S, Leirisalo-Repo M, Huitfeldt B, Juhlin R, Veys E, et al. a cross-sectional comparative study. Arthritis Care Res 2013;65:969-76. Sulfasalazine in the treatment of spondylarthropathy; a randomized, multicenter, double- blind, placebo-controlled study. Arthritis Rheum 1995;38:618-27. [26] Van Eijk IC, de Vries MK, Levels JHM, Peters MJL, Huizer EE, Dijkmans BAC, et al. Improvement of lipid profile is accompanied by atheroprotective alterations in high- [13] Clegg DO, Reda DJ, Abdellatif M. Comparison of sulfasalazine and placebo for 9 density lipoprotein composition upon tumor necrosis factor blockade: a prospective the treatment of axial and peripheral articular manifestations of the seronegative cohort study in ankylosing spondylitis. Arthritis Rheum 2009;60:1324-30. spondylarthropathies: a Department of Veterans Affairs cooperative study. Arthritis Rheum 1999;42:2325-9. [27] Tam LS, Shang Q, Kun EW, Lee KL, Yip ML, Li M, et al. The effects of golimumab on subclinical atherosclerosis and arterial stiffness in ankylosing spondylitis - a randomized, [14] Chen J, Liu C. Sulfasalazine for ankylosing spondylitis. Cochrane Database Syst Rev 10 placebo-controlled pilot trial. Rheumatology(Oxford) 2014;53:1065-74. 2005;2:CD004800. [28] Van Sijl AM, van Eijk IC, Peters MJ, Serné EH, van der Horst-Bruinsma IE, Smulders [15] Zochling J, van der Heijde D, Dougados M, Braun J. Current evidence for the YM, et al. Tumour necrosis factor blocking agents and progression of subclinical management of ankylosing spondylitis: a systematic literature review for the ASAS/ atherosclerosis in patients with ankylosing spondylitis. Ann Rheum Dis 2015;74:119-23. EULAR management recommendations in ankylosing spondylitis. Ann Rheum Dis 2006;65:423-32.

152 153 Chapter 10 General discussion and summary

[29] Scott DL, Smolen JS, Kalden JR, van de Putte LB, Larsen A, Kvien TK, et al. Treatment [44] Callhoff J, Sieper J, Weiß A, Zink A, Listing J. Efficacy of TNFα blockers in patients with of active rheumatoid arthritis with leflunomide: two year follow up of a double blind, ankylosing spondylitis and non-radiographic axial spondyloarthritis: a meta-analysis. placebo controlled trial versus sulfasalazine. Ann Rheum Dis 2001;60:913-23. Ann Rheum Dis 2014 Apr 9. doi: 10.1136/annrheumdis-2014-205322. [Epub ahead of print] [30] Kaltwasser JP, Nash P, Gladman D, Rosen CF, Behrens F, Jones P, et al. Efficacy and safety [45] Van der Heijde D, Sieper J, Maksymowych WP, Dougados M, Burgos-Vargas R, Landewé of leflunomide in the treatment of psoriatic arthritis and psoriasis: a multinational, R, et al. 2010 Update of the international ASAS recommendations for the use of anti-TNF double-blind, randomized, placebo-controlled clinical trial. Arthritis Rheum 2004;50:1939- agents in patients with axial spondyloarthritis. Ann Rheum Dis 2011;70:905-8. 1 50. [46] Barkham N, Kong KO, Tennant A, Fraser A, Hensor E, Keenan AM, et al. The unmet [31] Haibel H, Rudwaleit M, Braun J, Sieper J. Six months open label trial of leflunomide in need for anti-tumour necrosis factor (anti-TNF) therapy in ankylosing spondylitis. active ankylosing spondylitis. Ann Rheum Dis 2005;64:124-6. Rheumatology (Oxford) 2005;44:1277-81. [32] Vandooren B, Kruithof E, Yu DT, Rihl M, Gu J, De Rycke L, et al. Involvement of matrix [47] Van der Cruyssen B, Ribbens C, Boonen A, Mielants H, de Vlam K, Lenaerts J, et al. The 2 metalloproteinases and their inhibitors in peripheral synovitis and down-regulation epidemiology of ankylosing spondylitis and the commencement of anti-TNF therapy in by tumor necrosis factor alpha blockade in spondylarthropathy. Arthritis Rheum daily rheumatology practice. Ann Rheum Dis 2007;66:1072-7. 2004;50:2942-53. [48] Stichting Farmaceutische Kengetallen. Data en feiten 2012. Het jaar 2011 in cijfers. Den [33] Maugars Y, Mathis C, Berthelot J-M, Charlier C, Prost A. Assessment of the efficacy of Haag, 2012. 3 sacroiliac corticosteroid injections in spondylarthropathies: a double-blind study. Br J [49] Park W, Hrycaj P, Jeka S, Kovalenko V, Lysenko G, Miranda P, et al. A randomised, double- Rheumatol 1996;35:767-70. blind, multicentre, parallel-group, prospective study comparing the pharmacokinetics, [34] Haibel H, Fendler C, Listing J, Callhoff J, Braun J, Sieper J. Efficacy of oral prednisolone safety, and efficacy of CT-P13 and innovator infliximab in patients with ankylosing in active ankylosing spondylitis: results of a double-blind, randomised, placebo- spondylitis: the PLANETAS study. Ann Rheum Dis 2013;72:1605-12. 4 controlled short-term trial. Ann Rheum Dis 2014;73:243-6. [50] Wolbink GJ, Vis M, Lems WF, Voskuyl AE, de Groot E, Nurmohamed MT, et al. [35] Chen J, Veras MM, Liu C, Lin J. Methotrexate for ankylosing spondylitis. Cochrane Development of antiinfliximab antibodies and relationship to clinical response in patients Database Syst Rev 2013;2:CD004524. with rheumatoid arthritis. Arthritis Rheum 2006;54:711-5. 5 [36] Braun J, van der Heijde D, Doyle MK, Han C, Deodhar A, Inman R, et al. Improvement [51] Plasencia C, Pascual-Salcedo D, Nuño L, Bonilla G, Villalba A, Peiteado D, et al. Influence in hemoglobin levels in patients with ankylosing spondylitis treated with infliximab. of immunogenicity on the efficacy of longterm treatment of spondyloarthritis with Arthritis Rheum 2009;61:1032-6. infliximab. Ann Rheum Dis 2012;71:1955-60. [37] Rudwaleit M, Gooch K, Michel B, Herold M, Thörner A, Wong R, et al. Adalimumab [52] De Vries MK, Brouwer E, van der Horst-Bruinsma IE, Spoorenberg A, van Denderen JC, 6 improves sleep and sleep quality in patients with active ankylosing spondylitis. J Jamnitski A, et al. Decreased clinical response to adalimumab in ankylosing spondylitis Rheumatol 2011;38:79-86. is associated with antibody formation. Ann Rheum Dis 2009;68:1787-8. [38] Braun J, Baraliakos X, Hermann KG, van der Heijde D, Inman RD, Deodhar AA,et al. [53] Krieckaert CL, Nurmohamed MT, Wolbink GJ. Methotrexate reduces immunogenicity Golimumab reduces spinal inflammation in ankylosing spondylitis: MRI results of the in adalimumab treated rheumatoid arthritis patients in a dose dependent manner. Ann 7 randomised, placebo- controlled GO-RAISE study. Ann Rheum Dis 2012;71:878-84. Rheum Dis 2012;71:1914-5. [39] Van den Bosch F, Kruithof E, Baeten D, Herssens A, de Keyser F, Mielants H, et al. [54] Pérez-Guijo VC, Cravo AR, Castro M del C, Font P, Muñoz-Gomariz E, Collantes-Estevez E. Randomized double-blind comparison of chimeric monoclonal antibody to tumor Increased efficacy of infliximab associated with methtrexate in ankylosing spondylitis. necrosis factor alpha (infliximab) versus placebo in active spondylarthropathy. Arthritis Joint Bone Spine 2007;74:254-8. 8 Rheum 2002;46:755-65. [55] Cantini F, Niccoli L, Benucci M, Chindamo D, Nannini C, Olivieri I, et al. Switching from [40] Gorman JD, Sack KE, Davis JC Jr. Treatment of ankylosing spondylitis by inhibition of infliximab to once-weekly administration of 50 mg etanercept in resistant or intolerant tumour necrosis factor alpha. N Engl J Med 2002;346:1349-56. patients with ankylosing spondylitis: results of a fifty-four-week study.Arthritis Rheum [41] Van der Heijde D, Kivitz A, Schiff MH, Sieper J, Dijkmans BAC, Braun J, et al. Efficacy and 2006;55:812-6. 9 safety of adalimumab in patients with ankylosing spondylitis: results of a multicenter. [56] Lie E, van der Heijde D, Uhlig T, Mikkelsen K, Rødevand E, Koldingsnes W, et al. Randomised, double-blind, placebo-controlled trial. Arthritis Rheum 2006;54:2136-46. Effectiveness of switching between TNF inhibitors in ankylosing spondylitis: data from [42] Inman RD, Davis JC Jr, van der Heijde D, Diekman L, Sieper J, Kim SI, et al. Efficacy and the NOR-DMARD register. Ann Rheum Dis 2011;70:157-63. safety of golimumab in patients with ankylosing spondylitis: results of a randomized, [57] Glintborg B, Østergaard M, Krogh NS, Tarp U, Manilo N, Loft AG, et al. Clinical response, 10 double-blind, placebo-controlled, phase III trial. Arthritis Rheum 2008;58:3402-12. drug survival and predictors thereof in 432 ankylosing spondylitis patients after [43] Landewé R, Braun J, Deodhar A, Dougados M, Maksymowych WP, Mease PJ, et al. switching tumour necrosis factor α inhibitor therapy: results from the Danish nationwide Efficacy of certolizumab pegol on signs and symptoms of axial spondyloarthritis DANBIO registry. Ann Rheum Dis 2013;72:1149-55. including ankylosing spondylitis: 24-week results of a double-blind randomised placebo- controlled phase 3 study. Ann Rheum Dis 2014;73:39-47.

154 155 Chapter 10 General discussion and summary

[58] Ramiro S, Gaujoux-Viala C, Nam JL, Smolen JS, Buch M, Gossec L, et al. Safety of [72] Van Weely SF, van Denderen JC, Steultjens MP, van der Leeden M, Nurmohamed synthetic and biological DMARDs: a systematic literature review informing the 2013 MT, Dekker J, et al. Moving instead of asking? Performance-based tests and BASFI- update of the EULAR recommendations for management of rheumatoid arthritis. Ann questionnaire measure different aspects of physical function in ankylosing spondylitis. Rheum Dis 2014;73:529-35. Arthritis Res Ther 2012;14:R52. [59] Burmester GR, Panaccione R, Gordon KB, McIlraith MJ, Lacerda AP. Adalimumab: long- [73] Terwee CB, van der Slikke RM, van Lummel RC, Benink RJ, Meijers WG, de Vet HC. Self- term safety in 23 458 patients from global clinical trials in rheumatoid arthritis, juvenile reported physical functioning was more influenced by pain than performance-based 1 idiopathic arthritis, ankylosing spondylitis, psoriatic arthritis, psoriasis and Crohn’s physical functioning in knee-osteoarthritis patients. J Clin Epidemiol 2006;59:724-31. disease. Ann Rheum Dis 2013;72:517-24. [74] Niedermann K, Sidelnikov E, Muggli C, Dagfinrud H, Hermann M, Tamborrini G, et al. [60] Fouque-Aubert A, Jette-Paulin L, Combescure C, Basch A, Tebib J, Gossec L. Serious Effect of cardiovascular training on fitness and perceived disease activity in people with infections in patients with ankylosing spondylitis with and without TNF blockers: a ankylosing spondylitis. Arthritis Care Res 2013;65:1844-52. 2 systematic review and meta-analysis of randomised placebo-controlled trials. Ann [75] Bethi S, Dasgupta A, Weisman MH, Learch TJ, Gensler LS, Davis JC Jr, et al. Functional Rheum Dis 2012;71:1434. limitations due to axial and peripheral joint impairments in patients with ankylosing [61] Westhovens I, Lories RJ, Westhovens R, Verschueren P, de Vlam K. Anti-TNF therapy spondylitis: are focused measures more informative? Arthritis Care Res 2013;65:607-14. and malignancy in spondyloarthritis in the Leuven spondyloarthritis biologics cohort [76] Van Genderen S, van den Borne C, Geusens P, van der Linden S, Boonen A, Plasqui G. 3 (BIOSPAR). Clin Exp Rheumatol 2014;32:71-6. Physical functioning in patients with ankylosing spondylitis: comparing approaches of [62] Smitten AL, Simon TA, Hochberg MC, Suissa S. A meta-analysis of the incidence of experienced ability with self-reported and objectively measured physical activity. J Clin malignancy in adult patients with rheumatoid arthritis. Arthritis Res Ther 2008;10:R45. Rheumatol 2014;20:133-7. [63] Askling J, Baecklund E, Granath F, Geborek P, Fored M, Backlin C, et al. Anti-tumour [77] Kiltz U, Baraliakos X, Karakostas P, Igelmann M, Kalthoff L, Klink C, et al. The degree of 4 necrosis factor therapy in rheumatoid arthritis and risk of malignant lymphomas: relative spinal inflammation is similar in patients with axial spondyloarthritis who report high or risks and time trends in the Swedish Biologics Register. Ann Rheum Dis 2009;68:648-53. low levels of disease activity: a cohort study. Ann Rheum Dis 2012;71:1207-11. [64] Chakravarty EF, Michaud K, Wolfe F. Skin cancer, rheumatoid arthritis, and tumor necrosis [78] Machado P, Landewé RB, Braun J, Baraliakos X, Hermann KG, Hsu B, et al. MRI factor inhibitors. J Rheumatol 2005;32:2130-5. inflammation and its relation with measures of clinical disease activity and different 5 [65] Sun LM, Muo CH, Liang JA, Chang SN, Sung FC, Kao CH. Increased risk of cancer for treatment responses in patients with ankylosing spondylitis treated with a tumour patients with ankylosing spondylitis: a nationwide population-based retrospective cohort necrosis factor inhibitor. Ann Rheum Dis 2012;71:2002-5. study. Scand J Rheumatol 2014;43:301-6. [79] Elewaut D, Matucci-Cerinic M. Treatment of ankylosing spondylitis and extra-articular 6 [66] Hellgren K, Smedby KE, Backlin C, Sundstrom C, Feltelius N, Eriksson JK, et al. manifestations in everyday rheumatology practice. Rheumatology (Oxford) 2009;48:1029- Ankylosing spondylitis, psoriatic arthritis, and risk of malignant lymphoma: a cohort 35. study based on nationwide prospectively recorded data from Sweden. Arthritis [80] Zeboulon M, Dougados M, Gossec L. Prevalence and characteristics of uveitis in the Rheumatol 2014;66:1282-90. spondyloarthropathies: a systemic literature review. Ann Rheum Dis 2008;67:955-9. 7 [67] Chu CJ, Lu RH, Wang SS, Chang FY, Wu SL, Lu CL, et al. Risk factors associated with [81] Van der Horst-Bruinsma IE, Nurmohamed MT. Management and evaluation of extra- non-alcoholic fatty liver disease in Chinese patients and the role of tumor necrosis factor- articular manifestations in spondyloarthritis. Ther Adv Musculoskelet Dis 2012;4:413-22. alpha. Hepatogastroenterology 2007;54:2099-102. [82] Tubach F, Salmon D, Ravaud P, Allanore Y, Goupille P, Bréban M, et al. Risk of tuberculosis [68] Van der Heijde D, Dijkmans B, Geusens P, Sieper J, DeWoody K, Williamson P, et al. is higher with anti-tumor necrosis factor monoclonal antibody therapy than with soluble 8 Efficacy and safety of infliximab in patients with ankylosing spondylitis: results of a tumor necrosis factor receptor therapy: The three-year prospective French Research Axed randomized, placebo-controlled trial (ASSERT). Arthritis Rheum 2005;52:582-91. on Tolerance of Biotherapies registry. Arthritis Rheum 2009;60:1884-94. [69] Calin A, Jones SD, Garrett SL, Kennedy LG. Bath Ankylosing Spondylitis Functional [83] Cohen JD, Bournerias I, Buffard V, Paufler A, Chevalier X, Bagot M, et al.Psoriasis Index. Br J Rheumatol 1995;34:793-4. induced by tumor necrosis factor-alpha antagonist therapy: a case series. J Rheumatol 9 [70] Van der Heijde D, Calin A, Dougados M, Khan MA, van der Linden S, Bellamy N. 2007;34:380-5. Selection of instruments in the core set for DC-ART, SMARD, physical therapy, and [84] Fouache D, Goëb V, Massy-Guillemant N, Avenel G, Bacquet-Deschryver H, Kozyreff- clinical record keeping in ankylosing spondylitis. Progress report of the ASAS Working Meurice M, et al. Paradoxical adverse events of anti-tumour necrosis factor therapy for Group. Assessments in Ankylosing Spondylitis. J Rheumatol 1999;26:951-4. spondyloarthropathies: a retrospective study. Rheumatology (Oxford) 2009;48:761-4. 10 [71] Hoeymans N, Feskens EJ, van den Bos GA, Kromhout D. Measuring functional status: [85] Fiehn C, Vay S. Induction of inflammatory bowel disease flares by golimumab: report of cross-sectional and longitudinal associations between performance and self-report three patients with enteropathic spondylarthritis or ankylosing spondylitis and comorbid (Zutphen Elderly Study 1990-1993). J Clin Epidemiol 1996;49:1103-10. colitis. Arthritis Rheum 2011;63:3640-1. [86] Visvanathan S, van der Heijde D, Deodhar A, Wagner C, Baker DG, Han J, et al. Effects of infliximab on markers of inflammation and bone turnover and associations with bone mineral density in patients with ankylosing spondylitis. Ann Rheum Dis 2009;68:175-82.

156 157 Chapter 10 General discussion and summary

[87] Arends S, Spoorenberg A, Houtman PM, Leijsma MK, Bos R, Kallenberg CG, et al. The [101] Song IH, Hermann KG, Haibel H, Althoff CE, Listing J, Burmester GR, et al. Effects of effect of three years of TNFα blocking therapy on markers of bone turnover and their etanercept versus sulfasalazine in early axial spondyloarthritis on active inflammatory predictive value for treatment discontinuation in patients with ankylosing spondylitis: a lesions as detected by whole-body MRI (ESTHER): a 48-week randomised controlled trial. prospective longitudinal observational cohort study. Arthritis Res Ther 2012;14:R98. Ann Rheum Dis 2011;70:590-6. [88] Maksymowych WP, Jhangri GS, Fitzgerald AA, LeClercq S, Chiu P, Yan A, et al. A six- [102] Sieper J, van der Heijde D, Dougados M, Mease P, Maksymowych WP, Brown MA, month randomized, controlled, double-blind, dose-response comparison of intravenous et al. Efficacy and safety of adalimumab in patients with non-radiographic axial 1 pamidronate (60 mg versus 10 mg) in the treatment of nonsteroidal antiinflammatory spondyloarthritis: results of a randomised placebo-controled trial (ABILITY-1). Ann drug-refractory ankylosing spondylitis. Arthritis Rheum 2002;46:766-73. Rheum Dis 2013;72:815-22. [89] Clunie GP, Ginawi A, O’Conner P, Bearcroft PW, Garber SJ, Bhagat S, et al. An open-label [103] Sieper J, Lenaerts J, Wollenhaupt J, Rudwaleit M, Mazurov VI, Myasoutova L, et al. study of zoledronic acid (Aclasta 5 mg iv) in the treatment of ankylosing spondylitis. Ann Efficacy and safety of infliximab plus naproxen versus naproxen alone in patients with 2 Rheum Dis 2014;73:1273-4. early, active axial spondyloarthritis: results from the double-blind, placebo-controlled [90] Viapiana O, Gatti D, Idolazzi L, Fracassi E, Adami S, Troplini S, et al. Bisphosphonates vs INFAST study, Part 1. Ann Rheum Dis 2014;73:101-7. infliximab in ankylosing spondylitis treatment. Rheumatology (Oxford) 2014;53:90-4. [104] Dougados M, van der Heijde D, Sieper J, Braun J, Maksymowych WP, Citera G, et al. [91] Maksymowych WP, Chiowchanwisawakit P, Clare T, Pedersen SJ, Østergaard M, Symptomatic efficacy of etanercept and its effects on objective signs of inflammation in 3 Lambert RG. Inflammatory lesions of the spine on magnetic resonance imaging predict early nonradiographic axial spondyloarthritis: a multicenter, randomized, double-blind, the development of new syndesmophytes in ankylosing spondylitis: evidence of a placebo-controlled trial. Arthritis Rheumatol 2014;66:2091-102. relationship between inflammation and new bone formation. Arthritis Rheum 2009;60:93- [105] Rudwaleit M, Haibel H, Baraliakos X, Listing J, Märker-Hermann E, Zeidler H, et al. The 102. early disease stage in axial spondylarthritis: results from the German Spondyloarthritis 4 [92] Van der Heijde D, Machado P, Braun J, Hermann KG, Baraliakos X, Hsu B, et al. MRI Inception Cohort. Arthritis Rheum 2009;60:717-27. inflammation at the vertebral unit only marginally predicts new syndesmophyte [106] Ciurea A, Scherer A, Exer P, Bernhard J, Dudler J, Beyeler B, et al. Tumor necrosis factor formation: a multilevel analysis in patients with ankylosing spondylitis. Ann Rheum Dis α inhibition in radiographic and nonradiographic axial spondyloarthritis: results from a 2012;71:369-73. large observational cohort. Arthritis Rheum 2013;65:3096-3106. 5 [93] Van der Heijde D, Landewé R, Baraliakos X, Houben H, van Tubergen A, Williamson P, [107] Van der Horst-Bruinsma IE. Treatment of non-radiographic axial spondyloarthritis: it is et al. Radiographic findings following two years of infliximab therapy in patients with only the beginning. Ann Rheum Dis 2013;72:789-90. ankylosing spondylitis. Arthritis Rheum 2008;58:3063-70. [108] Kiltz U, Baraliakos X, Karakostas P, Igelmann M, Kalthoff L, Klink C, et al. Do patients with [94] Van der Heijde D, Landewé R, Einstein S, Ory P, Vosse D, Ni L, et al. Radiographic non-radiographic axial spondylarthritis differ from patients with ankylosing spondylitis? 6 progression of ankylosing spondylitis after up to two years of treatment with etanercept. Arthritis Care Res 2012;64:1415-22. Arthritis Rheum 2008;58:1324-31. [109] Wallis D, Haroon N, Ayearst R, Carty A, Inman RD. Ankylosing spondylitis and [95] Van der Heijde D, Salonen D, Weissman BN, Landewé R, Maksymowych WP, Kupper H, nonradiographic axial spondyloarthritis: part of a common spectrum or distinct diseases? et al. Assessment of radiographic progression in the spines of patients with ankylosing J Rheumatol 2013;40:2038-41. 7 spondylitis treated with adalimumab for up to 2 years. Arthritis Res Ther 2009;11:R127. [110] Moltó A, Paternotte S, Claudepierre P, Breban M, Dougados M. Effectiveness of tumor [96] Maksymowych WP, Elewaut D, Schett G. Motion for debate: the development of ankylosis necrosis factor α blockers in early axial spondyloarthritis: data from the DESIR cohort. in ankylosing spondylitis is largely dependent on inflammation. Arthritis Rheum Arthritis Rheumatol 2014;66:1734-44. 8 2012;64:1713-9. [111] Davis JC Jr, van der Heijde DM, Braun J, Dougados M, Clegg DO, Kivitz AJ, et al. [97] Haroon N, Inman RD, Learch TJ, Weisman MH, Lee M, Rahbar MH, et al. The impact of Efficacy and safety of up to 192 weeks of etanercept therapy in patients with ankylosing tumor necrosis factor α inhibitors on radiographic progression in ankylosing spondylitis. spondylitis. Ann Rheum Dis 2008;67:346-52. Arthritis Rheum 2013;65:2645-54. [112] Baraliakos X, Listing J, Fritz C, Haibel H, Alten R, Burmester GR, et al. Persistent clinical 9 [98] Baraliakos X, Haibel H, Listing J, Sieper J, Braun J. Continuous long-term anti-TNF efficacy and safety of infliximab in ankylosing spondylitis after 8 years: early clinical therapy does not lead to an increase in the rate of new bone formation over 8 years in response predicts long-term outcome. Rheumatology (Oxford) 2011;50:1690-9. patients with ankylosing spondylitis. Ann Rheum Dis 2014;73:710-5. [113] Sieper J, van der Heijde D, Dougados M, Brown LS, Lavie F, Pangan AL. Early response [99] Braun J, Baraliakos X, Hermann KG, Deodhar A, van der Heijde D, Inman R, et al. The to adalimumab predicts long-term remission through 5 years of treatment in patients 10 effect of two golimumab doses on radiographic progression in ankylosing spondylitis: with ankylosing spondylitis. Ann Rheum Dis 2012;71:700-6. results through 4 years of the GO-RAISE trial. Ann Rheum Dis 2014;73:1107-13. [114] Breban M, Vignon E, Claudepierre P, Devauchelle V, Wendling D, Lespessailles E, et al. [100] Spoorenberg A, de Vlam K, van der Linden S, Dougados M, Mielants H, van de Tempel H, Efficacy of infliximab in refractory ankylosing spondylitis: results of a six-month open- et al. Radiological scoring methods in ankylosing spondylitis. Reliability and change over label study. Rheumatology (Oxford) 2002;41:1280-5. 1 and 2 years. J Rheumatol 2004;31:125-32.

158 159 Chapter 10 General discussion and summary

[115] Brandt J, Khariouzov A, Listing J, Haibel H, Sörensen H, Grassnickel L, et al. Six-month [128] Ritchlin C, Rahman P, Kavanaugh A, McInnes IB, Puig L, Li S, et al. Efficacy and safety of results of a double-blind, placebo-controlled trial of etanercept treatment in patients with the anti-IL-12/23 p40 monoclonal antibody, ustekinumab, in patients with active psoriatic active ankylosing spondylitis. Arthritis Rheum 2003;48:1667-75. arthritis despite conventional non-biological and biological anti-tumour necrosis factor [116] Navarro-Compán V, Moreira V, Ariza-Ariza R, Hernández-Cruz B, Vargas-Lebrón C, therapy: 6-month and 1-year results of the phase 3, multicentre, double-blind, placebo- Navarro-Sarabia F. Low doses of etanercept can be effective in ankylosing spondylitis controlled, randomised PSUMMIT 2 trial. Ann Rheum Dis 2014;73:990-9. patients who achieve remission of the disease. Clin Rheumatol 2011;30:993-6. [129] Poddubnyy D, Hermann K-GA, Callhoff J, Listing J, Sieper J. Ustekinumab for 1 [117] De Stefano R, Frati E, De Quattro D, Menza L, Manganelli S. Low doses of etanercept can the treatment of patients with active ankylosing spondylitis: results of a 28-week, be effective to maintain remission in ankylosing spondylitis patients. Clin Rheumatol prospective, open-label, proof-of-concept study (TOPAS). Ann Rheum Dis 2014;73:817-23. 2014;33:707-11. [130] Benham H, Rehaume LM, Hasnain SZ, Velasco J, Baillet AC, Ruutu M, et al. Interleukin-23 [118] Song IH, Althoff CE, Haibel H, Hermann KG, Poddubnyy D, Listing J, et al. Frequency mediates the intestinal response to microbial β-1,3-glucan and the development of 2 and duration of drug-free remission after 1 year of treatment with etanercept versus spondyloarthritis pathology in SKG mice. Arthritis Rheumatol 2014;66:1755-67. sulfasalazine in early axial spondyloarthritis: 2 year data of the ESTHER trial. Ann Rheum [131] Noordenbos T, Yeremenko N, Gofita I, Van der Sande M, Tak PP, Caňete JD, et al. Dis 2012;71:1212-5. Interleukin-17-positive mast cells contribute to synovial inflammation in spondylarthritis. [119] Haibel H, Heldmann F, Braun J, Listing J, Kupper H, Sieper J. Long-term efficacy Arthritis Rheum 2012;64:99-109. 3 of adalimumab after drug withdrawal and retreatment in patients with active non- [132] Baeten D, Baraliakos X, Braun J, Sieper J, Emery P, van der Heijde D, et al. Anti- radiographically evident axial spondyloarthritis who experience a flare.Arthritis Rheum interleukine-17A monoclonal antibody secukinumab in treatment of ankylosing spondylitis: 2013;65:2211-3. a randomised, double-blind, placebo-controlled trial. Lancet 2013;382:1705-13. [120] Sieper J, Lenaerts J, Wollenhaupt J, Rudwaleit M, Mazurov VI, Myasoutova L, et al. 4 Maintenance of biologic-free remission with naproxen or no treatment in patients with early, active axial spondyloarthritis: results from a 6-month, randomised, open-label follow-up study, INFAST Part 2. Ann Rheum Dis 2014;73:108-13. [121] Brandt J, Listing J, Haibel H, Sörensen H, Schwebig A, Rudwaleit M, et al. Long-term 5 efficacy and safety of etanercept after readministration in patients with active ankylosing spondylitis. Rheumatology (Oxford) 2005;44:342-8. [122] Haibel H, Rudwaleit M, Listing J, Sieper J. Open label trial of anakinra in active ankylosing spondylitis over 24 weeks. Ann Rheum Dis 2005;64:296-8. 6 [123] Song IH, Heldmann F, Rudwaleit M, Haibel H, Weiss A, Braun J, et al. Treatment of active ankylosing spondylitis with abatacept: an open-label, 24-week pilot study. Ann Rheum Dis 2011;70:1108-10. 7 [124] Song IH, Heldmann F, Rudwaleit M, Listing J, Appel H, Braun J, et al. Different response to rituximab in tumor necrosis factor blocker-naive patients with active ankylosing spondylitis and in patients in whom tumor necrosis factor blockers have failed: a twenty- four-week clinical trial. Arthritis Rheum 2010;62:1290-7. 8 [125] Sieper J, Porter-Brown B, Thompson L, Harari O, Dougados M. Assessment of short-term symptomatic efficacy of tocilizumab in ankylosing spondylitis: results of randomised, placebo-controlled trials. Ann Rheum Dis 2014;73:95-100. [126] Sieper J, Braun J, Kay J, Badalamenti S, Radin AR, Jiao L, et al. Sarilumab for the 9 treatment of ankylosing spondylitis: results of a Phase II, randomised, double- blind, placebo-controlled study (ALIGN). Ann Rheum Dis 2014 Feb 18. doi: 10.1136/ annrheumdis-2013-204963. [Epub ahead of print] [127] Pathan E, Abraham S, Van Rossen E, Withrington R, Keat A, Charles PJ, et al. Efficacy and 10 safety of apremilast, an oral phosphodiesterase 4 inhibitor, in ankylosing spondylitis. Ann Rheum Dis 2013;72:1475-80.

160 161 Chapter 10 Nederlandse samenvatting

2. NEDERLANDSE SAMENVATTING de ontdekking van aspirine in de negentiende eeuw zijn ontstekingsremmende middelen de belangrijkste therapie voor AS. Met name de niet-steroide anti- Inleiding inflammatoire medicijnen (NSAIDs) zijn nog steeds de hoeksteen van de Het onderwerp van dit proefschrift is de medicamenteuze behandeling van de behandeling. Traditioneel wordt verondersteld dat deze middelen vooral de aandoening ankyloserende spondylitis (AS). In Nederland wordt AS nog vaak symptomen onderdrukken. 1 “de ziekte van Bechterew” genoemd. Bij patiënten met AS is er sprake van een chronische reumatische ontsteking in gewrichten, vooral in het bekken en de Daarnaast spreekt men in de reumatologie van ziekte modificerende medicijnen wervelkolom. De oorzaak van de ontsteking is niet precies bekend, maar bij het (DMARDs) die geacht worden het beloop van de ziekte te beïnvloeden. Besproken ontstaan van de ziekte spelen erfelijke aanleg, bacteriën in de darm en waarschijnlijk wordt dat dit onderscheid echter niet altijd zo zwart-wit is. Dit proefschrift bevat in de 2 andere factoren van buiten een belangrijke rol. De ontsteking veroorzaakt rugpijn, eerste hoofdstukken studies waarin het effect van DMARDs bij AS is onderzocht. met name aan het eind van de nacht, ochtendstijfheid, moeheid en beperkingen in het dagelijkse functioneren. Kenmerkend voor AS is de aangroei van bot die kan Sedert circa 2002 zijn andere type medicijnen in gebruik bij AS: de zogenaamde 3 optreden als gevolg van de ontstekingen. Hierdoor kunnen blijvende verstijvingen biologicals. Deze bestaan uit eiwitten die een onderdrukking bewerkstelligen van van de wervelkolom ontstaan en in een ver gevorderd stadium de kenmerkende specifieke signaaleiwitten die een rol spelen bij ontstekingen. Deze zogenaamde voorovergebogen houding. TNF-alpha blokkers kunnen alleen per injectie of infuus worden toegediend en blijken een heel gunstig effect op de symptomen van AS te kunnen hebben. De 4 Historie laatste hoofdstukken van dit proefschrift betreffen studies waarin diverse aspecten In de inleiding van het proefschrift wordt beschreven dat er kenmerken van AS van behandeling met TNF-blokkers zijn onderzocht. zijn gevonden in oude skeletten, zelfs al in die van Egyptische farao’s. Toch werd lange tijd AS niet als een aparte ziekte gezien, maar als een vorm van acuut reuma Aan het eind van het proefschrift wordt een overzicht gegeven van de 5 of gewrichtsreuma. In de loop van de negentiende eeuw werden wel diverse medicamenteuze mogelijkheden bij AS zoals die zich in de afgelopen vijftien jaar ziektegevallen gepubliceerd die passen bij AS. Aan het eind van die eeuw begon hebben ontwikkeld. Hieronder volgt daarvan de samenvatting. men AS als een aparte aandoening te beschouwen na publicaties van onder andere 6 de Russische neuroloog W. Bechterew. Toch duurde het nog meer dan een halve Niet-steroide anti-inflammatoire medicijnen (NSAIDs) eeuw tot de ziekte in de hele wereld als zodanig werd erkend. Ontwikkelingen die NSAIDs zijn nog steeds de geneesmiddelen van eerste keus bij de behandeling van hier ook aan hebben bijgedragen, waren röntgen – en bloedonderzoek (met name AS vanwege hun gunstige werking op pijn en stijfheid. Voor de meeste patiënten bepaling van erfelijke factor HLA-B27). is andere medicamenteuze behandeling niet noodzakelijk. Uiteraard moeten 7 de voordelen van behandeling altijd worden afgewogen tegen de mogelijke In 1963 zijn internationale criteria voor classificatie van AS opgesteld die in 1984 bijwerkingen. Hoewel het risico op maagzweren bij nieuwe soorten NSAIDs zijn gewijzigd en nog steeds worden gebruikt. De belangrijkste criteria zijn: rugpijn lager is, kunnen deze middelen nadelig werken op bloeddruk, hart, nieren en de 8 passend bij ontsteking, bewegingsbeperking van rug en borstkas en een röntgenfoto bloedstolling. met kenmerken van ontsteking. Men zag al jaren dat er bij AS overeenkomsten en ook overlap was met andere reumatische ontstekingen, zoals bij de huidziekte NSAIDs lijken echter meer te doen dan alleen bestrijding van de symptomen. psoriasis, bij chronische darmontstekingen en reactieve gewrichtsontsteking Jaren geleden werd in een Nederlandse studie al gevonden dat continu gebruik 9 na bepaalde infecties. De overkoepelende term voor deze ziektebeelden werd van een NSAID ook de overmatige aangroei van bot bij AS afremt. Recente spondylarthropathie en later (axiale) spondyloartritis (SpA). Recent zijn hiervoor studies hebben deze bevinding bevestigd, met name bij patiënten met verhoogde nieuwe criteria opgesteld met inbegrip van het vroege stadium van AS waarin ontstekingswaarden in het bloed. Dit leidt tot de aanbeveling om patiënten met 10 röntgenfoto’s nog niet afwijkend zijn. progressieve ziekte en verhoogde ontstekingswaarden chronisch te behandelen met een NSAID. Therapie Patiënten met AS worden reeds lang effectief behandeld met fysiotherapie en in de afgelopen tien jaar is het nut van oefentherapie in onderzoeken bevestigd. Sinds

162 163 Chapter 10 Nederlandse samenvatting

Ziekte modificerende medicijnen (DMARDs) TNF-blokkers Sulfasalazine heeft gunstige effecten op de ziekte AS indien er niet uitsluitend Er zijn inmiddels meerdere TNF-blokkers beschikbaar met potentieel zeer gunstige ontstekingen in de rug zijn, maar tevens gewrichtsontstekingen in gewrichten buiten effecten op verschijnselen van AS. Internationale criteria zijn opgesteld voor de de rug (perifeer). Sulfasalazine wordt in de darm gesplitst in het antibacteriële indicaties van deze behandeling. Bij de inzetbaarheid van deze behandeling moet sulfapyridine en het ontstekingsremmende mesalazine. Mesalazine is het werkzame rekening gehouden worden met zowel mogelijke bijwerkingen als hoge kosten. 1 bestanddeel bij de behandeling van chronische darmontstekingen. Het effect van mesalazine op AS werd in een open studie gedurende 24 weken onderzocht bij Inmiddels is gevonden dat het menselijk lichaam antistoffen kan maken tegen een 20 patiënten met actieve ziekte (hoofdstuk 2). Veel patiënten moesten voortijdig aantal TNF-blokkers met als gevolg bijwerkingen of een verminderde werking. Dit stoppen met de behandeling vanwege bijwerkingen. Hoewel de gemiddelde werd ook onderzocht bij 38 AS patiënten die gedurende 54 weken met de TNF- 2 ontstekingswaarde in het bloed significant daalde, veranderden andere blokker infliximab werden behandeld (hoofdstuk 5). Bij de patiënten waarbij het ziekteparameters niet gedurende de behandeling met mesalazine. medicijn niet goed werkte, werden significant vaker antistoffen tegen het medicijn gevonden en daardoor ook significant lagere bloedspiegels van de TNF-blokker. Of Cholesterol verlagende medicijnen hebben ook een ontstekingsremmende werking. antistofvorming geremd kan worden door het gelijktijdig geven van methotrexaat, is 3 Het effect van de cholesterolverlager rosuvastatine werd daarom onderzocht bij bij AS nog niet goed onderzocht. De bepalingen van bloedspiegels van het medicijn 15 patiënten met AS in een open studie gedurende 12 weken met een observatie en de antistoffen daartegen, zouden een rol kunnen gaan spelen bij beslissingen periode van totaal 24 weken (hoofdstuk 3). Ook hier werd een significante daling van zoals stoppen of wijziging van de behandeling met een TNF-blokker. Hiervoor is 4 de gemiddelde ontstekingswaarde in het bloed gevonden, maar toonden klinische echter nog verder onderzoek nodig. parameters alleen een trend tot verbetering. De waarden van het cholesterol in het bloed verbeterden. Hetgeen gunstig is, omdat het risico op hart- en vaatziekten door De belangrijkste bijwerking van TNF-blokkers is het verhoogde risico op een ernstige een reumatische ontsteking op zich ook al verhoogd is. infectie. Dit risico lijkt bij AS kleiner dan bij andere reumatische aandoeningen; 5 mogelijk omdat bij AS minder vaak andere weerstandsverlagende medicatie wordt Leflunomide is een DMARD met bewezen effect bij andere reumatische ziekten. bijgegeven. Vanaf de introductie van TNF-blokkers is er ook bezorgdheid geweest Het effect van leflunomide werd onderzocht in een dubbelblinde gerandomiseerde over een mogelijke toename van ontwikkeling van kwaadaardige aandoeningen. Er 6 studie bij 45 patiënten met actieve AS (hoofdstuk 4). De resultaten van de patiënten zijn weinig studies met grote aantallen AS patiënten, maar daarin was het risico op behandeld met leflunomide en die met een placebo verschilden niet significant. In maligniteiten niet verhoogd. een Duitse studie werden aanwijzingen gevonden dat het middel wel een gunstig Bij de behandeling met een TNF-blokker zijn bij AS afwijkingen beschreven van effect had bij patiënten met perifere gewrichtsontsteking. Dit konden wij niet de leverwaarden in het bloed. Dit hebben we onderzocht bij 105 AS patiënten die 7 onderzoeken, omdat perifere gewrichtsontsteking bij onze patiënten bijna niet minstens drie maanden waren behandeld met de TNF-blokker etanercept (hoofdstuk voorkwam. 6). Bij 15 patiënten werd minstens tweemaal een leverwaarde gevonden hoger dan 150% van de bovengrens van normaal. In negen gevallen was er waarschijnlijk of 8 Andere DMARDs, zoals we die kennen bij gewrichtsreuma, zijn bij AS relatief weinig mogelijk een relatie met het medicijn. Dit bleek vaker voor te komen bij te zware onderzocht. Recent is een studie gepubliceerd waarin een korte termijn effect werd personen. Aanbevolen wordt daarom gedurende het eerste jaar van de behandeling gezien van een kortdurende behandeling met een hoge dosis prednison. Het effect met een TNF-blokker geregeld de leverwaarden in het bloed te controleren. van methotrexaat is enkele keren onderzocht met negatief resultaat, maar steeds in 9 lagere doseringen dan we gewend zijn te gebruiken bij gewrichtsreuma. Het effect van TNF-blokkers wordt vooral gemeten door middel van internationaal getoetste vragenlijsten. Het is echter bekend dat de uitkomsten van vragenlijsten Concluderend is de rol van traditionele DMARDs bij de behandeling van AS nog kunnen verschillen van testen die op een andere wijze de activiteit en gevolgen 10 steeds beperkt. Er is een rol voor sulfasalazine en misschien voor leflunomide bij van de ziekte meten. In onze studie werden de uitkomsten van een vragenlijst AS met perifere gewrichtsontsteking. Er is geen rol voor mesalazine. De rol van vergeleken met de uitkomsten van fysieke functietesten bij 82 patiënten voor en na cholesterolverlagers, methotrexaat in hogere doseringen en ook prednison zou behandeling gedurende 3 maanden met een TNF-blokker (hoofdstuk 7). Het bleek dat verder onderzocht kunnen worden in placebo gecontroleerde studies. een fors aantal patiënten enerzijds niet goed reageerden op het medicijn volgens internationale criteria en volgens de uitkomsten van vragenlijsten, maar anderzijds

164 165 Chapter 10 Nederlandse samenvatting

wel beter functioneerden tijdens de uitvoering van fysieke functietesten. Hieruit Zo is er de vraag of patiënten in een zo vroeg mogelijk stadium met TNF-blokkers bleek dat de verschillende testen deels andere aspecten van het functioneren meten. moeten worden behandeld. Meerdere TNF-blokkers zijn nu ook geregistreerd voor Aanbevolen wordt het effect van medicatie op meerdere manieren te meten en de axiale spondyloarthritis wanneer er (nog) geen afwijkingen op de röntgenfoto verschillen en overlap tussen deze methoden verder te onderzoeken. zichtbaar zijn. In deze gevallen moet alleen behandeling met een TNF-blokker overwogen worden als er wel kenmerken zijn van ontsteking in het bloed of op 1 Bij AS komen niet alleen ontstekingen voor in rug en perifere gewrichten, maar afbeeldingen van een MRI scan. Daarover is internationaal consensus. ook van de huid, darm of oog, de zogenoemde extra-articulaire verschijnselen. Verder is niet bekend hoe lang behandeling met een TNF-blokker moet worden Ontsteking van het regenboogvlies (uveitis anterior) komt het meeste voor; bij voortgezet. Waarschijnlijk vlamt de ziekte bij een groot deel van de AS patiënten circa 33% van de AS patiënten treedt deze oogontsteking minimaal eenmaal op. op na stoppen van de behandeling. Anderzijds kan bij een flink aantal van de 2 Het effect van de TNF-blokker adalimumab op het voorkomen van deze uveitis werd behandelingen de dosis van het medicijn worden verminderd zonder verlies van onderzocht bij 77 patiënten die gedurende minstens 12 weken waren behandeld effectiviteit. Ook blijkt hervatten van de behandeling na staken over het algemeen vanwege actieve AS in de rug (hoofdstuk 8). Ten opzichte van het jaar voor de weer succesvol. behandeling nam het aantal aanvallen van uveitis significant af. Mede op grond Ondanks het succes van TNF-blokkers in het algemeen bij AS, blijft er een grote 3 van andere onderzoeken kan geconcludeerd worden dat TNF-blokkers een gunstige groep AS patiënten bestaan waarbij TNF-blokkers niet werkzaam zijn of waarbij de werking hebben op extra-articulaire verschijnselen, maar dat er tussen de TNF- behandeling gestopt moet worden vanwege bijwerkingen. Er blijft dus behoefte aan blokkers wat dat betreft wel onderlinge verschillen zijn. nieuwe medicijnen. Diverse geneesmiddelen, die werkzaam zijn bij gewrichtsreuma, 4 blijken niet werkzaam te zijn bij AS. Maar de eerste resultaten met een eiwit, dat Bij AS bestaat de bijzondere combinatie van enerzijds plaatselijke overmatige in plaats van TNF-alpha een andere ontstekingfactor (interleukine-17) remt, zijn botvorming en anderzijds een afname van de algehele botmassa (osteoporose) veelbelovend. met verhoogde kans op botbreuken (onder andere wervelinzakkingen). De invloed 5 op bot van behandeling met de TNF-blokker etanercept gedurende twee jaar werd Diverse suggesties voor verder onderzoek zijn beschreven aan het eind van de onderzocht bij 49 patiënten met AS (hoofdstuk 9). Het bleek dat de botmassa discussie. tijdens behandeling was toegenomen. Anderzijds nam het aantal inzakkingen van 6 wervels eveneens toe en nam ook de voor AS kenmerkende plaatselijke overmatige Concluderende opmerking botvorming toe. Aangezien een onbehandelde controle groep ontbrak, kan niet Er is in de afgelopen 15 jaar een enorme positieve ontwikkeling op gang gekomen worden uitgesloten dat een TNF-blokker op deze laatste fenomenen toch een in de medicamenteuze behandeling van AS. Verder onderzoek is echter nodig om te remmend effect heeft. kunnen blijven werken aan het uiteindelijke doel: verbetering van het welzijn en de 7 Andere medicijnen, die algemeen gebruikt worden bij osteoporose, zijn nog ziekte-uitkomst van patiënten met de ziekte van Bechterew. nauwelijks onderzocht bij AS. Bij verder onderzoek is niet alleen het effect op de botmassa, maar vooral op het optreden van botbreuken van belang. 8

Wat betreft de voor AS kenmerkende plaatselijke overmatige botvorming, waren ook de resultaten van andere onderzoekers, die de toename van botvorming vergeleken met historische controle groepen, teleurstellend. Het leek alsof de TNF-blokkers, 9 ondanks zeer gunstige effecten op ontsteking, deze botvorming niet afremden en in die zin dus geen ziekte modificerende werking hadden. Recentere onderzoeken suggereren echter dat weliswaar na het tot rust komen van ontstekingen plaatselijke 10 botaangroei plaatsvindt, maar dat op termijn van vele jaren tijdens behandeling met een TNF-blokker deze botgroei toch minder optreedt.

Tenslotte worden een aantal actuele thema’s betreffende de behandeling met TNF- blokkers besproken.

166 167 Chapter 10 Dankwoord en Curriculum vitae

3. DANKWOORD 4. CURRICULUM VITAE

Alle personen die op enige wijze hebben bijgedragen aan de totstandkoming van dit De schrijver van dit proefschrift werd geboren op 9 april 1957 te Kampen. In 1975 proefschrift wil ik hierbij heel hartelijk bedanken. werd het eindexamen atheneum B behaald aan het Johannes Calvijn Lyceum aldaar. De geneeskunde studie aan de Universiteit Utrecht werd in maart 1982 1 De dankbaarheid is groot, maar ik houd het graag bij een simpele opsomming van afgerond. Ter vervulling van de vervangende dienstplicht werkte hij daarna als de mensen die hebben geholpen. AGNIO reumatologie in het Gemeente Ziekenhuis Arnhem bij de reumatologen J.W. Boersma, AP Hartman en J Weber. Van juli 1984 tot september 1987 was hij Zij die inmiddels niet meer onder ons zijn: Betty Dekker-Saeys, Marcel van der Paardt AGIO interne geneeskunde in het Diaconessenhuis te Leeuwarden (opleider M.P. 2 en Rob van de Stadt. Leemhuis) en van september 1987 tot juli 1989 AGIO reumatologie in het Gemeente Hans Bijlsma, mijn promotor. Ziekenhuis Arnhem (opleider J.W. Boersma). Irene en Mike, mijn vasthoudende copromotoren. 25 jaar nadat in 1964 de eerste paal was geslagen van het Jan van Breemen Instituut 3 Ben Dijkmans, de initiator. (JBI) aan de Dr Jan van Breemenstraat te Amsterdam, begon hij daar zijn werk als De promotiecommissie. reumatoloog en inmiddels is hij daar 25 jaar werkzaam. Namens het JBI werkt hij Mijn paranymfen. twee dagen per week in het Waterlandziekenhuis te Purmerend. De mede-onderzoekers, met name Mike Peters, Mirjam de Vries, Salima van Weely, Reade is de nieuwe naam van de organisatie sinds de fusie in 2010 van het JBI 4 Maria Suttorp-Schulten, Ingrid Visman en Mignon van der Weijden. met het revalidatiecentrum Amsterdam aan de Overtoom. In 2014 is het reuma Mijn werkgever en de afdeling wetenschappelijk onderzoek. onderzoek van AMC, Reade en VUmc gebundeld in het Amsterdam Rheumatology & Mijn ex- en huidige collega reumatologen en artsen. Center (ARC). De patiënten, die meewerkten aan studies en om wiens welzijn het allemaal draait. 5 De (reuma)verpleegkundigen, die patiënten begeleiden en veel data verzamelen. Aaf, Inez en al hun collega’s, die de spreekuren en de dokter ondersteunen. Ellen, voor de omslag en layout. 6 Mijn gezin. 7

8

9

10

168 169 Chapter 10 List of publications

5. LIST OF PUBLICATIONS 12. VP van Halm, JC van Denderen, MJL Peters, JWR Twisk, M van der Paardt, IE van der Horst-Bruinsma, RJ van de Stadt, MHMT de Koning, BAC Dijkmans, MT Nurmohamed. Increased disease activity is associated with a deteriorated lipid profile in patients with 1. JC van Denderen, JW Boersma, P Zeinstra, AP Hollander, BR van Neerbos. Physiological ankylosing spondylitis. Ann Rheum Dis 2006;65:1473-1477. effects of exhaustive physical exercise in primary syndrome (PFS): is PFS a disorder of neuroendocrine reactivity? Scand J Rheumatol 1992;21:35-7. 13. MK de Vries, GJ Wolbink, SO Stapel, H de Vrieze, JC van Denderen, BAC Dijkmans, LA Aarden, IE van der Horst-Bruinsma. Decreased clinical response to infliximab in 2. M Boers, AC Verhoeven, HM Markusse, MAFJ van de Laar, R Westhovens, JC van 1 ankylosing spondylitis is correlated with anti-infliximab formation. Ann Rheum Dis Denderen, D van Zeben, BAC Dijkmans, AJ Peeters, P Jacobs, HR van den Brink, HJ 2007;66:1252-4. Schouten, DMFM van der Heijde, A Boonen, S van der Linden. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with suphasalazine 14. D van Schaardenburg, WH Bos, J Ursum, RJ van de Stadt, JC van Denderen, BAC alone in early rheumatoid artritis. Lancet 1997;350:309-18. Dijkmans. De diagnostische accuratesse van de anticitrullinebepaling voor het vaststellen 2 van reumatoïde artritis door de huisarts. Ned Tijdschr Geneeskd 2008;152:1244. 3. JC van Denderen, EAM Mensen, AM Vints. Artritis door gebruik van rifabutine bij Mycobacterium avium-infectie. Ned Tijdschr Geneesk 1997;141:2028-30. 15. MK de Vries, IE van der Horst-Bruinsma, MT Nurmohamed, LA Aarden, SO Stapel, MJ Peters, JC van Denderen, BAC Dijkmans, GJ Wolbink. Immunogenicity does not influence 4. M van der Paardt, JC van Denderen, AJC van den Brule, SA Morré, IE van der Horst- treatment with etanercept in patients with ankylosing spondylitis. Ann Rheum Dis Bruinsma, PD Bezemer, BAC Dijkmans. Prevalence of Chlamydia trachomatis in urine of 3 2009;68:531-5. male patients with ankylosing spondylitis is not increased. Ann Rheum Dis 2000;59:300-2. 16. SFE van Weely, JC van Denderen, IE van der Horst-Bruinsma, MT Nurmohamed, BAC 5. EJ Giltay, C Popp-Snijders, JC van Denderen, D van Schaardenburg, LJG Gooren, BAC Dijkmans, J Dekker, MP Steultjens. Reproducibility of performance measures of physical Dijkmans. Phenylbutazone can spuriously elevate unextracted testosterone assay results function based on the BASFI, in ankylosing spondylitis. Rheumatology 2009;48:1254-60. in patients with ankylosing spondylitis. Letter to the editor. J Clin Endocrinol Metab 4 2000;85:4923-4. 1 7. MK de Vries, E Brouwer, IE van der Horst-Bruinsma, A Spoorenberg, JC van Denderen, A Jamnitski, MT Nurmohamed, BAC Dijkmans, LA Aarden, GJ Wolbink. Decreased clinical 6. AE van Ede, RFJM Laan, MJ Rood, TWJ Huizinga, MAFJ van de Laar, JC van Denderen, response to adalimumab in ankylosing spondylitis is associated with antibody formation. AAA Westgeest, TC Romme, DJRAM de Rooij, MJM Jacobs, TM de Boo, GJ van der Wilt, Ann Rheum Dis 2009;68:1787-8. JL Severens, M Hartman, PFM Krabbe, BAC Dijkmans, FC Breedveld, LBA van de Putte. 5 Effect of folic or folinic acid supplementation on the toxicity and efficacy of methotrexate 18. LH van Tuyl, M Boers, WF Lems, RBM Landewé, H Han, S van der Linden, MAFJ van de in rheumatoid arthritis. Arthritis Rheum 2001;44:1515-24. Laar, R Westhovens, JC van Denderen, ML Westedt, AJ Peeters, P Jacobs, TWJ Huizinga, HR van den Brink, BAC Dijkmans, AE Voskuyl. Survival, comorbidities and joint damage 11 7. RBM Landewé, M Boers, AC Verhoeven, R Westhovens, MAFJ van der Laar, HM Markusse, years after the COBRA combination therapy trial in early rheumatoid arthritis. Ann Rheum 6 JC van Denderen, ML Westedt, AJ Peeters, BAC Dijkmans, P Jacobs, A Boonen, DMFM Dis 2010;69:807-12. van der Heijde, S van der Linden. COBRA combination therapy in patients with early rheumatoid arthritis: long-term structural benefits of a brief intervention. Arthritis Rheum 19. MK de Vries, AS van Drumpt, BJ van Royen, JC van Denderen, RA Manoliu, IE van der 2002;46:347-56. Horst-Bruinsma. Discovertebral (Andersson) lesions in severe ankylosing spondylitis: a study using MRI and conventional radiography. Clin Rheumatol 2010;29:1433-8. 7 8. AE van Ede, RFJM Laan, MJ Rood, TWJ Huizinga, MAFJ van de Laar, JC van Denderen, AAA Westgeest, TC Romme, DJRAM de Rooij, MJM Jacobs, TM de Boo, GJ van der Wilt, 20. MAC van der Weijden, JC van Denderen, WF Lems, MW Heymans, BAC Dijkmans, IE van JL Severens, M Hartman, PFM Krabbe, BAC Dijkmans, FC Breedveld, LBA van de Putte. der Horst-Bruinsma. Low bone mineral density is related to male gender and decreased Het effect van comedicatie met foliumzuur of folinezuur op toxiciteit en effectiviteit functional capacity in early spondylarthropathies. Clin Rheumatol 2011;30:497-503. van methotrexaat bij de behandeling van reumatoïde artritis: een gerandomiseerde, 21. SFE van Weely, JC van Denderen, MP Steultjens, M van der Leeden, MT Nurmohamed, 8 dubbelblinde, placebogecontroleerde studie gedurende 48 weken. Ned Tijdschr Geneesk J Dekker, BAC Dijkmans, IE van der Horst-Bruinsma. Moving instead of asking? 2002;146:1322-27. Performance-based tests and BASFI-questionnaire measure different aspects of physical 9. JC van Denderen, IE van der Horst-Bruinsma, PD Bezemer, BAC Dijkmans. Efficacy and function in ankylosing spondylitis. Arthritis Res Ther 2012;14(2);R52. 9 safety of Mesalazine (Salofalk) in an open study of 20 patients with ankylosing spondylitis. 22. ST Bruijnen, MAC van der Weijden, JP Klein, OS Hoekstra, R Boellaard, JC van Denderen, J Rheumatol 2003;30:1558-60. BAC Dijkmans, AE Voskuyl, IE van der Horst-Bruinsma, CJ van der Laken. Bone formation 10. JC van Denderen, M van der Paardt, MT Nurmohamed, YM de Ryck, BAC Dijkmans, IE van rather than inflammation reflects Ankylosing Spondylitis activity on PET-CT: a pilot study. der Horst-Bruinsma. Double blind, randomised, placebo controlled study of leflunomide in Arthritis Res Ther 2012; 14(2):R71. 10 the treatment of active ankylosing spondylitis. Ann Rheum Dis 2005;64:1761-4. 23. MAC van der Weijden, IE van der Horst-Bruinsma, JC van Denderen, BAC Dijkmans, MW 11. JC van Denderen, MJL Peters, VP van Halm, IE van der Horst-Bruinsma, BAC Dijkmans, Heymans, WF Lems. High frequency of vertebral fractures in early spondylarthropathies. MT Nurmohamed. Statin therapy might be beneficial for treating patients with ankylosing Osteoporos Int 2012;23:1683-90. spondylitis. Ann Rheum Dis 2006;65:695-696. 24. JC van Denderen, GJ Blom, IE van der Horst-Bruinsma, BAC Dijkmans, MT Nurmohamed. Elevated liver enzymes in patients with ankylosing spondylitis treated with etanercept. Clin Rheumatol 2012;31:1677-82.

170 171 Chapter 10 List of abbreviations and illustrations

25. SFE van Weely, JC van Denderen, MP Steultjens, MT Nurmohamed, BAC Dijkmans, J 6. LIST OF ABBREVATIONS AND ILLUSTRATIONS Dekker, IE van der Horst-Bruinsma. What do we miss? ASAS non-responders on anti- TNF therapy show improvement in performance-based physical function. Rheumatology ADA adalimumab (Oxford). 2013;52:1884-9. AE adverse events 26. JC van Denderen, IM Visman, MT Nurmohamed, MSA Suttorp-Schulten, IE van der Horst- Bruinsma. Adalimumab significantly reduces the recurrence rate of anterior uveitis in aHR adjusted hazard ratio 1 patients with Ankylosing Spondylitis. J Rheumatol 2014;41:1843-8. ALT alanine aminotransferase 27. SFE van Weely, J Dekker, MP Steultjens, JC van Denderen, MT Nurmohamed, BAC ANA antinuclear antibody Dijkmans, IE van der Horst-Bruinsma. Objective evaluation of physical functioning after AS ankylosing spondylitis TNFi therapy in Ankylosing Spondylitis patients: a selection of three feasible performance- 2 5-ASA 5-aminosalicyl acid based tests. J Rheumatol 2015 Jan 15. pii: jrheum.140337. [Epub ahead of print] ASAS(20) Assessment of SpondyloArthritis international Society (20%); formerly: ASsess- ment in Ankylosing Spondylitis ASDAS ankylosing spondylitis disease activity score 3 ASQOL ankylosing spondylitis quality of live AST aspartate aminotransferase AU anterior uveitis 4 BASDAI Bath ankylosing spondylitis disease activity index BASFI Bath ankylosing spondylitis functional index BASG Bath ankylosing spondylitis global score BASMI Bath ankylosing spondylitis metrology index 5 BC before Christ BMD bone mineral density BMI body mass index 6 CI confidence interval COXIB cyclooxygenase-2 selective inhibitor CRP C-reactive protein 7 CT computer tomography CTX-I,II carboxy terminal crosslinking telopeptide type I,II collagen DEXA dual-energy X-ray absorptiometry DISH diffuse idiopathic skeletal hyperostosis (Forestier’s disease) 8 DMARD disease modifying antirheumatic drug DNA deoxyribonucleic acid ESR erythrocyte sedimentation 9 ELISA enzyme-linked immunosorbent assay ETN etanercept GDA global disease activity 10 GWB D/P general wellbeing according to doctor / patient HBsAg hepatitis B surface antigen HDL high density lipoprotein HF heart frequency HLA human leukocyte antigen

172 173 Chapter 10 List of abbreviations and illustrations

IBD inflamatory bowel disease Verantwoording illustraties ICC intraclass correlation coefficient Omslag: Sassenpoort Zwolle, gebouwd in 1406-09 (foto auteur). Introductie: IFX infliximab Figuur 1: Slide collection 7-C-3, American College of Rheumatology. IQR interquartile range Figuur 2: D O’Connell. Ankylosing spondylitis: the literature up to the close of the IL interleukin 1 nineteenth century. Ann Rheum Dis 1956;15:119-23. INH isoniazide Figuur 3: L Disqué: Naturgemäße Behandlung der Krankheiten. Leipzig, 1904. LDL low density lipoprotein Figuur 4, 5: A Goldscheider und P Jacob: Handbuch der physikalischen Therapie. MRI magnetic resonance imaging Verlag von George Thieme, Leipzig, 1902. 2 MMP matrix metalloproteinase Figuur 6: JMH Moll, ed. Ankylosing spondylitis. Churchill Livingstone, Edinburgh / mSASSS modified Stoke ankylosing spondylitis spine score New York, 1980. MTX methotrexate Figuur 7: J Glax: Lehrbuch der Balneotherapie. Verlag von Ferdinand Enke, Stuttgart, NAFLD non-alcoholic fatty liver disease 1897. 3 NSAID nonsteroidal anti-inflammatory drug Figuur 8, 9, 10: FE Bilz: De nieuwe natuurgeneeswijze. Karel Meijer & Co, OL open-label Amsterdam, 1905. OPG osteoprotegerin Figuur 11: RKW Kuipers. De orthopaedische behandeling van spondylarthritis 4 PCR polymerase chain reaction ankylopoetica. Ned Tijdschr Geneeskd 1947;91:851-6. PP per-protocol Advertenties: E Dietrich, M Hirsch, red.: Rheuma-Jahrbuch. Leo Alterthum Verlag, RA rheumatoid arthritis Berlin-Charlottenburg, 1929 en 1930/31. 5 RANKL receptor activator of nuclear factor kappa-B ligand RCT randomized controlled trial SD standard deviation SJS / SJC swollen joint score / count 6 SpA spondyloarthropathy / spondyloarthritis SSZ sulfasalazine TJS tender joint score 7 TNF tumour necrosis factor-alpha UNL upper normal limit VAS visual analogue scale 8 VF vertebral fractures X-rays röntgen radiation 9

10

174 175