Duration of etanercept treatment and reasons for discontinuation in a ...

Rheumatology 2011;50:189–195 doi:10.1093/rheumatology/keq308 Advance Access publication 3 November 2010

RHEUMATOLOGY

Original article Duration of etanercept treatment and reasons for discontinuation in a cohort of juvenile idiopathic arthritis patients Taunton R. Southwood1, Helen E. Foster2, Joyce E. Davidson3, Kimme L. Hyrich4, Catherine B. Cotter5, Lucy R. Wedderburn6, Richard G. Hull7, Helen E. Venning8, Joy K. Rahman1 and Carole L. Cummins1 on behalf of the British Society for Adolescent and Paediatric Rheumatology Biologics and New Drugs Register Abstract Objective. Since 2004, juvenile idiopathic arthritis (JIA) patients treated with etanercept and/or MTX have been monitored in the British Society for Paediatric and Adolescent Rheumatology Biologics and New Drug Register. Here, we report the duration of etanercept use for the first 5 years of the register and reasons for discontinuation.

Results. A total of 483 etanercept-treated JIA patients were enrolled from 30 UK centres, representing 941 patient-years of follow-up. A total of 100 (20.7%) patients discontinued etanercept; 9 due to disease control, 88 because of treatment failure, 2 for unknown reasons and 1 because of a change in diagnosis. Of the 53 patients in whom etanercept was perceived to be ineffective at controlling the inflammation, 48 were prescribed other biologic drugs [26/48 (54%) infliximab]. In 21 patients with intolerance, infections, CNS events and a few isolated events were associated with discontinuation. Using Kaplan–Meier analysis, at 5 years 69% (95% CI 61, 77%) had not experienced treatment failure. Discontinuation of etanercept for inefficacy was associated with systemic arthritis subtype [odds ratio (OR) 2.55, 95% CI 1.27, 5.14], chronic anterior uveitis (OR 2.39, 95% CI 1.06, 5.35) and inefficacy of MTX before starting etanercept (OR 8.3, 95% CI 1.14, 60.58). Conclusions. In a cohort of JIA patients treated with etanercept and followed for a median of 2 years (maximum 5 years), the majority (69%) remain on the drug. Key words: Juvenile idiopathic arthritis, Child, Arthritis, Etanercept, Methotrexate, Serious adverse event, Drug discontinuations, Disease remission. 1 Department of Paediatric Rheumatology, Institute of Child Health, Birmingham Children’s Hospital – NHS Foundation Trust and University of Birmingham, Birmingham, 2Musculoskeletal Research Group, Institute Cellular Medicine, Newcastle University and Newcastle Hospitals NHS Foundation Trust, Newcastle upon Tyne, 3 Rheumatology Department, Royal Hospital for Sick Children, Yorkhill, Glasgow, 4Arthritis Research UK Epidemiology Unit, University of Manchester, Manchester, 5Institute of Child Health, Birmingham Children’s Hospital – NHS Foundation Trust and University of Birmingham, Birmingham, 6Institute of Child Health, University College London and Great Ormond Street Hospital, London, 7Department of Rheumatology, Queen Alexandra Hospital, Portsmouth and 8 Nottingham Children’s and Young Peoples Rheumatology Service, Nottingham Children’s Hospital, Nottingham, UK.

Submitted 6 April 2010; revised version accepted 18 August 2010. Correspondence to: Taunton R. Southwood, Institute of Child Health, 4th Floor Laboratories Block, Birmingham Children’s Hospital – NHS Foundation Trust, Whittall Street, Birmingham, B4 6NH, UK. E-mail: [email protected]

Introduction Juvenile idiopathic arthritis (JIA) is one of the most common chronic inflammatory diseases of childhood. It is characterized by persistent synovial inflammation for which a wide variety of drugs are used, either sequentially or in combination, including NSAIDs, corticosteroids (CSs), MTX and a wide variety of biologic agents [1]. The efficacy of most of these drugs in controlling synovial inflammation has been demonstrated in clinical trials of selected JIA populations. MTX treatment has been established in paediatric rheumatology for >20 years [2]. When arthritis activity is not controlled with MTX or MTX is not tolerated, the next therapeutic step is usually addition

! The Author 2010. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: [email protected]

CLINICAL SCIENCE

Methods. Disease subtype and activity, comorbidity, treatment efficacy and safety data were recorded. Etanercept discontinuation was defined as stopping the drug because of disease remission or treatment failure. Time to discontinuation was explored using Kaplan–Meier survival analysis with remaining patients censored at 5-year follow-up.

Taunton R. Southwood et al.

of a biologic agent such as a TNF-a inhibitor, either alone or in combination with MTX [3]. The number of children and young people with JIA treated with biological agents has increased rapidly since the first randomized controlled trial of a TNF-a inhibitor (etanercept) was published [4]. The original trial patients have been followed for up to 8 years to assess longer term safety and efficacy [5]. In order to monitor the ‘real world’ safety and efficacy in JIA, etanercept registries have been developed in many countries and more than 2500 JIA patients have been followed for an estimated 5000 patient-years [6–10]. There have been few reports of etanercept discontinuation rates and factors associated with discontinuing the drug [6, 9]. The aim of this article was to explore the reasons for etanercept discontinuation in a national cohort of JIA patients.

Materials and methods The British Society for Paediatric and Adolescent Rheumatology (BSPAR) initiated a national Biologics and New Drugs Registry (BNDR) in 2004. The consent of parents and patients (as appropriate) was obtained according to the Declaration of Helsinki. The study was approved by the West Midlands Research Ethics Committee. All patients fulfilled the ILAR criteria for the classification of JIA [11]. The use of etanercept in the UK is limited to those patients who are intolerant to MTX, have arthritis that is not controlled by MTX, or both [12]. The aim of the registry is to collect long-term outcome data on children receiving etanercept for JIA. A second cohort of biologically naive children with JIA treated with MTX is also being recruited with which to compare outcomes. After informed consent, patient data could be collected retrospectively if the patient began etanercept or MTX before enrolment into the BSPAR BNDR, or prospectively started at the time of enrolment, or both. In either case, data were collected from the point at which etanercept (former cohort) or MTX (latter cohort) was started. The data reported in this article are from the etanercept-treated cohort only. Data were contributed by 30 UK centres after they were given local research ethics committee approval (listed in the ‘Acknowledgements’ section). A standardized set of baseline data in five domains was collected from each patient, including details of patient demography, JIA classification (ILAR criteria), JIA disease activity including the ACR core data set [13], comorbidity and concurrent medication. Follow-up data in the same five domains were collected at pre-determined intervals after commencing the registered drug; 3 months, 6 months, 1 year and annually thereafter. Adverse event data was requested as part of monitoring, but adverse events could also be reported on an ad hoc basis. Each centre had the option to contribute standardized data in paper format or electronically via a Web-based data entry system developed specifically for the BSPAR BNDR. Any decisions to start and discontinue etanercept were made locally by the patient’s rheumatology team. Discontinuation was defined as cessation of the drug because of disease control, apparent intolerance

190

(including lack of adherence and stopping associated with adverse events) or physician-perceived inefficacy. Patients in whom etanercept was stopped temporarily for a planned event such as surgery were not included in the discontinuation group. An adverse event was defined as any untoward medical occurrence that occurred during treatment with a registered drug, and a serious adverse event (SAE) was defined according to modified medicines and healthcare products regulatory agency (MHRA) and international conference on harmonisation (ICH) guidelines [14]. That is, a SAE was any untoward medical occurrence that, at any dose, resulted in death, a life-threatening illness, hospitalization, persistent or significant disability or incapacity, a congenital anomaly or birth defect, or a medically significant event determined by the responsible local physician that in their opinion jeopardized the health of the patient and required intervention to prevent one of the other outcomes listed.

Statistical analysis Descriptive statistics were used for demographic data. Kaplan–Meier survival analysis was used to determine time to drug discontinuation for inefficacy, intolerance or non-adherence with patients censored at 5-year follow-up, if adequate disease control allowing drug discontinuation was obtained, and in one case because the diagnosis of JIA was revised to Crohn’s diseaseassociated arthritis. A Cox regression model developed using backward stepwise models was used to highlight possible associations between variables and outcome measures, with the choice of variables for inclusion informed by inspection of the correlation matrix to allow for multicolinearity.

Results During the first 4 years of operation, 483 etanercepttreated JIA patients were enrolled in the BSPAR Registry from 30 UK centres; 275 (57%) of these patients were also treated with MTX (Table 1). Ninety-one per cent had monitoring data [median duration of follow-up: 24 months (range 1–90 months)], representing 941 patient-years of follow-up. The patient demographics (gender distribution, JIA subtypes and disease duration) reflected the more severe end of the JIA spectrum, as might be expected of a JIA cohort that was unresponsive or intolerant to MTX; 67% of patients were female, and 16% had systemic arthritis (and these were more likely to be treated with both etanercept and MTX; P = 0.009). The median age at starting etanercept was 12.0 years (range 2–21 years, interquartile range 6.0). In total, 100 (20.7%) patients discontinued etanercept; 9 due to disease control (duration of benefit 3–56 months) and 88 because of treatment failure [53 (11%) due to inefficacy, 14 (2.9%) due to non-compliance and 21 (4.3%) due to adverse events] (Table 2). In the remaining three patients, the reasons for discontinuation were not known in two and the diagnosis of one patient was changed to IBD-related arthritis. Among all treatment failures, combination MTX and etanercept users were

www.rheumatology.oxfordjournals.org

Discontinuation of etanercept in JIA

TABLE 1 Patient characteristics at entry to the register

Patient characteristics Total etanercept registered patients Female White Reasons for starting etanercept Inadequate response to MTX Intolerant of MTX Inadequate response and intolerance Systemic arthritis Polyarthritis RF+ Polyarthritis RF Oligoarthritis extended Oligoarthritis persistent PsA Enthesis-related arthritis Undifferentiated arthritis Unspecified Ever had chronic anterior uveitis Concurrent medications at baseline Oral CSs MTX SSZ Ciclosporin HCQ LEF Previous medications MTX SSZ Ciclosporin IVIG CYC Oral CSs i.v. CSs IA CSs

n (%) n = 483 325 (67) 417 (86) 275 85 116 77 48 157 79 11 30 38 36 7 54

(57) (18) (24) (16) (9) (33) (16) (2) (6) (8) (7) (1) (11)

184 279 14 9 8 3

(38) (57) (3) (2) (2) (0.4)

483 67 35 14 4 339 210 385

(100) (13.9) (7.2) (2.9) (0.8) (70.2) (43.5) (79.7)

Median (S.D.) Age, years Disease duration, years

12 (3.6) 5 (3.4)

TABLE 2 Reasons for etanercept discontinuation in 100 patients

Outcome

Patient number (per cent of treatment failure)

Treatment success Treatment failurea Adverse events, n (%) Inefficacy, n (%) Non-adherence, n (%) No information Total

9 88 21 (24%) 53 (60%) 14 (16%) 3a 100

a

One patient changed from JIA to IBD-associated arthritis, so censored in the survival analysis at that point.


TABLE 3 Etanercept discontinuations due to disease control

JIA subtype

Time to Stopped, discontinuation, n Restarted min

Systemic arthritis

4

ERA

2

Polyarthritis RF Extended oligoarthritis Undifferentiated Total

No No No Yes

56 12 3 12

1 1

No Yes Yes No

3 34 10 11

1 9

Yes 4

5

more likely to fail for inefficacy compared with users of etanercept alone. Of those patients in whom etanercept was discontinued for inefficacy, 15 were taking etanercept alone at baseline (i.e. 45% of a total of 33 treatment failures on etanercept alone at baseline) and 38 were taking a combination of etanercept and MTX at baseline (i.e. 69% of a total of 55 treatment failures on both drugs at baseline; 2 Yates correction = 3.87, 1 degrees of freedom, P = 0.049). Of the nine patients who stopped etanercept because of disease control, four were restarted on it for increased disease activity at the discretion of the local paediatric rheumatology team as of the last recorded follow-up. The median duration of disease control before restarting etanercept was 15.25 months (range 5–34 months). Four patients (36%) had the systemic arthritis subtype, of whom only one restarted etanercept during the follow-up period (Table 3). The 21 patients who discontinued etanercept due to drug intolerance experienced a wide range of adverse events. Twenty-one adverse events associated with discontinuation fell into three groups: infections, noninfectious CNS events and other. Infections (Epstein–Barr meningoencephalitis, pyelonephritis, recurrent urinary tract infection, gingival infection, sepsis) were seen in five patients. Non-infectious CNS events included headaches (2), low mood (2), panic/anxiety attacks (1), hallucinations (1), optic neuritis (1), blurred vision (1), and uveitis flare (1). Other events were eczema (1), menorrhagia (1), low white cell count (1), diarrhoea (1), pregnancy (1), inadvertent measles, mumps and rubella vaccination (1) and macrophage activation syndrome (1). Two patients in the cohort died after discontinuation of etanercept use, one while admitted to hospital before bone marrow transplant and one after bone marrow transplant. Of those patients in whom etanercept was perceived as being ineffective at controlling the inflammation, 48/53 (90.5%) were prescribed other biologic drugs, mostly infliximab (26/48: 54%) (Table 4). In five cases, chronic anterior uveitis was cited as the reason the patients were

191


switched to inflixmab (two patients who were on etanercept alone, three who were also on MTX at baseline). A further four patients with a history of uveitis were switched to infliximab, although that was not stated as the reason for switching (one was on etanercept alone and three also on MTX at baseline). Five patients switched medication again, three of them back to etanercept. Using Kaplan–Meier analysis, at 1 year 91% (95% CI 88, 94%) had not discontinued treatment because of inefficacy, intolerance or TABLE 4 Changes in medication after stopping etanercept Switch to Infliximab

Adalimumab

n = 53

Further switches, n

26

Adalimumab, 1 Anakinra, 1 Rituximab, 1 Adalimumab then anakinra, 1 Adalimumab then rituximab, 1 Etanercept then adalimumab, 1 Etanercept then adalimumab then anakinra, 1 Etanercept, 1 Infliximab, 1 Adalimumab, 1 Etanercept then infliximab then anakinra None Etanercept

12

Anakinra LEF

7 1

Ciclosporin Clinical trial drug None

1 1 5

non-adherence. At 2 years, 84% (95% CI 80, 89%) were continuing, with 69% (95% CI 61, 77%) doing so at 5 years (18 patients remaining on follow-up) (Fig. 1). Further, in Kaplan–Meier analyses, discontinuation for these reasons at 5 years was not associated with gender, disease duration longer than 5 years, commencing etanercept before the age of 10 years, disease onset before the age of 5 years, or concurrent MTX or CS treatment. Cox regression models, exploring factors that might be associated with discontinuation for treatment failure (i.e. including inefficacy, intolerance or non-adherence) highlighted no associations. In Cox regression with backward stepwise model development, however, discontinuation of etanercept for physician-perceived inefficacy alone was associated with systemic arthritis subtype, chronic anterior uveitis and inefficacy of MTX before starting etanercept (Table 5).

Discussion The cohort of 483 JIA patients taking etanercept monitored in the BSPAR BNDR represents the more severe TABLE 5 Cox regression model of discontinuations of etanercept for inefficacy

Discontinuation for inefficacy alone Systemic arthritis Ever had chronic anterior uveitis Inefficacy of MTX before etanercept

Odds ratio (95% CI) 2.55 (1.27, 5.14) 2.39 (1.06, 5.35) 8.30 (1.14, 60.58)

FIG. 1 Duration of etanercept use over 5 years (median 2 years, range 1–90 months): discontinuations for inefficacy, intolerance and non-adherence. Survival function Censored

Cumulative continuation of etanercept

1.0 0.9 0.8 0.7

At 5 years: 69% (95% CI 61%, 77%)

0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

192

10

30 40 20 Time in months

50

60



end of the spectrum of JIA patients, those who had failed to tolerate or respond to conventional treatment strategies including MTX and a variety of NSAIDs, CS preparations and physical therapy. In the past, this type of patient would have received various combinations of slowacting anti-rheumatic drugs and would be unlikely to respond satisfactorily to any of them. Etanercept was introduced into the therapeutic armamentarium for JIA in 2000 after a single randomized withdrawal clinical trial [4] and widespread use in the UK followed publication of guidance from the National Institute of Clinical Excellence (NICE) [12]. The product licence for Enbrel in the UK mandated the collation of safety data from JIA in a national register (subsequently incorporated into BSPAR guidelines and recommendations from NICE) and charged prescribing consultants with the responsibility for registering JIA patients. There is relatively little information on the long-term use of etanercept in JIA cohorts; so the aim of the BSPAR BRND was to document the response of the enrolled JIA cohort over a 5-year period initially. Sixty-nine per cent of registered JIA patients treated with etanercept in the UK continued to take the drug at last clinic visit over a median follow-up period of 2 years, with the range of follow-up extending to 5 years. This is an encouraging outcome from a real-life clinical cohort consisting of JIA patients with the poorest prognostic outlook. Drug discontinuations provide relatively objective surrogate information for drug inefficacy, intolerance or both (none of which have been defined for paediatric populations who are not part of a randomized clinical trial). It is possible that continued use of etanercept may overestimate drug efficacy, but other criteria for defining treatment success, such as improvement in the ACR Core Data Set data, are difficult to collect comprehensively in real-life outpatient clinics. The prolonged use of etanercept suggests that the majority of this population find the drug tolerable and effective at controlling arthritis. Sustained benefit from etanercept has been demonstrated in 8-year follow-up data from the original patients enrolled in the randomized controlled trial [5] and several other national JIA etanercept registries. Other drug survival data have been reported by the Finnish registry (83%, 4-year drug survival), the German disease registry [83% (57/334) 1-year drug survival] and the Dutch registry [78% (33/146) 5-year drug survival] [6, 7, 9]. Direct comparison of drug survival data between registries is confounded by relatively small patient numbers and variation in the duration of follow-up. A smaller proportion of the UK cohort stopped taking etanercept because of apparent disease control [1% (9/483)] than that seen in the German cohort (4%) or the Finnish cohort (10%). This may reflect a shorter follow-up period or duration of etanercept treatment in many of our patients. The UK registry commenced enrolling patients more recently than most others [15–17], during a time when there was increasing availability of a wider variety of biologic agents; physicians may have preferred to discontinue etanercept and use other treatments rather than wait for a delayed response to etanercept. Alternatively, differences in our


enrolled patient population, such as a greater proportion of poor prognostic disease groups (compared with the German and Finnish data) or a more prolonged disease duration prior to treatment with etanercept (compared with the Dutch data), may have accounted for the lower frequency of drug discontinuation due to disease control. The most common reason for stopping etanercept in the UK cohort [53/100 (60%)] of patients who discontinued etanercept was physician perceived inadequate response to treatment, suggesting etanercept inefficacy. This represents 11% (53/483) of the UK etanercepttreated cohort, similar data to that reported from the Dutch cohort (14%: 20/146 patients) and the German cohort (9%: 65/722), but fewer than the Finnish cohort (28%: 29/105). It is interesting that etanercept discontinuations for inefficacy were proportionately greater in patients treated with a combination of MTX and etanercept than in users of etanercept alone. A possibility, which requires further study, is that the combination group included more patients whose disease was relatively resistant to all treatment, but in whom physicians were reluctant to stop MTX in the hope that it still had a marginal benefit. A wide range of adverse events were associated with etanercept discontinuation. Of particular concern was one patient who developed optic neuritis, a disease that has been reported previously in etanercept-treated JIA patients [18, 19]. Infection-associated adverse events have also been widely reported in JIA patients and adults with RA and accounted for a quarter of adverse events associated with drug discontinuation in the UK cohort [4–10, 20–24]. No cases of tuberculosis were documented in these UK patients who discontinued etanercept [20]. We only observed uveitis flare in one patient resulting in disease discontinuation, despite several reports associating etanercept with this complication of JIA [25–29]. One of the UK patients was withdrawn from the registry due to a change in the disease diagnosis from JIA to arthritis associated with IBD (Crohn’s disease), an association that has been postulated to be related to etanercept treatment [30, 31]. We did not find any discontinuations associated with the development of other chronic inflammatory or autoimmune diseases [32]. Menorrhagia has not been reported in adult patients with RA, but was seen in one UK JIA patient whose etanercept was discontinued [24]. No cases of malignancy were observed. About half of the UK cohort who discontinued etanercept (53/100 JIA patients, 11% of the total JIA cohort) were prescribed other biologic treatment, a somewhat smaller proportion than that observed in the Finnish cohort (35%), which was followed for 53 months [6]. The most frequently prescribed drugs were further TNF-a blocking agents (infliximab in 26 patients and adalimumab in 12 patients) or the IL-1 blocker anakinra. To investigate factors associated with discontinuing etanercept in the UK JIA cohort, we used Kaplan–Meier analyses and considered discontinuations for drug inefficacy, intolerance or non-adherence. None of these reasons for discontinuation were associated with

193


gender, disease duration longer than 5 years, commencing etanercept before age 10, disease onset before age 5 years, concurrent MTX or CS treatment. However, when Cox regression with backward stepwise modelling was developed, discontinuation of etanercept for inefficacy was associated with systemic arthritis subtype, chronic anterior uveitis and inefficacy of MTX before starting etanercept. Systemic arthritis has been proposed previously to be associated with a poorer response to etanercept [10, 33], as has uveitis, so these results were not unexpected. We also noted an association of ineffective etanercept responses in patients who had started therapy because they had found MTX to be ineffective. This suggests that poor therapeutic responder patients may be at higher risk of subsequent suboptimal responses to a change in treatment. In these patients, it is possible that factors perpetuating the chronic inflammatory process are independent of both TNF-a and the mechanisms of MTX action. Based on a time-to-event analysis in a cohort of JIA patients treated with etanercept and followed for up to 5 years (median 2 years), we conclude that the majority (69%) remained on the drug. Physician-perceived etanercept inefficacy, rather than adverse events, was the main reason for discontinuation. Rheumatology key messages Sixty-nine per cent etanercept-treated JIA patients remain on etanercept for a median of 2 years. . Four per cent (21/483) of patients stopped etanercept because of adverse events. . Systemic JIA, chronic anterior uveitis and MTX inefficacy were associated with etanercept inefficacy. .

Acknowledgements Thanks to Mrs Beverley Thomas for her untiring skills as Clinical Rheumatology Research Administration Coordinator. Thanks to all the clinicians who contributed to this study, including the participating centre coordinators at Addenbrookes Hospital, Cambridge (Andrew Ostor), Basildon & Thurrock University Hospital, Essex (Nagui Gendi), Birmingham Children’s Hospital NHS Foundation Trust (Tauny Southwood), Central Manchester University Hospital NHS Trust, Manchester (Alice Chieng), Chesterfield Royal Hospital NHS Trust (John Bourne), Derby City General Hospital, Derby (Richard Morton), Dumfries & Galloway Royal Infirmary, Scotland (Robert Simpson), George Eliot Hospital, University Hospitals of Coventry and Warwickshire NHS Trust (Kathy Bailey), Gloucestershire Hospitals NHS Foundation Trust, Cheltenham (Kate Martin), Great Ormond Street Hospital, London (Clarissa Pilkington), Hinchingbrooke Hospital, Huntingdon (Gill Pountain), Kings Mill Hospital, Mansfield (Jane Ellis), Leeds General Infirmary, Leeds (Sue Wyatt), Norfolk & Norwich University Hospital, Norwich (Kate Armon), Nuffield Orthopaedic Hospital, Oxford (Nick Wilkinson), Peterborough & Stamford Hospital NHS

194

Foundation Trust (Nick Sheehan), Poole Hospital, Dorset (Paul Thompson), Queen Alexandra Hospital, Portsmouth (Richard Hull), University College London, Middlesex (Debajit Sen), Queen Elizabeth, the Queen Mother Hospital, Margate (Alison Leak), Queen Elizabeth Hospital, Woolwich (Tim Price), Nottingham Children’s Hospital (Helen Venning), Royal Blackburn Hospital, Blackburn (Pat Smith), Royal Hospital for Sick Children, Glasgow and Edinburgh (Joyce Davidson), Royal Liverpool Children’s Hospital, Alder Hey, Liverpool (Liza McCann), Newcastle University and Newcastle Hospitals NHS Foundation Trust, Newcastle upon Tyne (Helen Foster), Royal Wolverhampton Hospitals NHS Trust, Wolverhampton (Karen Davies), Staffordshire Rheumatology Centre, Stafford (Jonathan Packham), University Hospital Bristol NHS Foundation Trust (Athimalaipet Ramanan), and University Hospital of Wales, Cardiff (Jeremy Camilleri). British Society for Paediatric and Adolescent Rheumatology Biologics and New Drug Register is funded by a grant from Wyeth Pharmaceuticals. Principal investigator was T.R.S. The national study coordinator (C.L.C.) received partial salary from the grant, the national nurse coordinator (C.B.C.) received full salary from the grant and the national data manager (J.K.R.) received full salary from the grant. Disclosure statement: R.G.H. was responsible for negotiating the initial funding of the register. His department has received educational grants for individuals to attend conferences and corporate grants to fund independent disease management conferences for people with arthritis and the public. All other authors have declared no conflicts of interest.

References 1 Ravelli A, Martini A. Juvenile idiopathic arthritis. Lancet 2007;369:767–78. 2 Wallace CA. The use of methotrexate in childhood rheumatic diseases. Arthritis Rheum 1998;4:381–91. 3 Schmeling H, Mathony K, John V, Keyaˆer G, Burdach S, Horneff G. A combination of etanercept and methotrexate for the treatment of refractory juvenile idiopathic arthritis: a pilot study. Ann Rheum Dis 2001;60:410–2. 4 Lovell DJ, Giannini EH, Reiff A et al. for the Pediatric Rheumatology Collaborative Study Group. Etanercept in children with polyarticular juvenile rheumatoid arthritis. N Engl J Med 2000;342:763–9. 5 Giannini EH, Ilowite NT, Lovell DJ et al. for the Pediatric Rheumatology Collaborative Study Group. Long-term safety and effectiveness of etanercept in children with selected categories of juvenile idiopathic arthritis. Arthritis Rheum 2009;60:2794–804. 6 Horneff G, Schmeling H, Biedermann T et al. The German etanercept registry for treatment of juvenile idiopathic arthritis. Ann Rheum Dis 2004;63:1638–44. 7 Prince FHM, Twilt M, Ten Cate R et al. Long-term follow-up on effectiveness and safety register of etanercept in juvenile idiopathic arthritis: the Dutch national register. Ann Rheum Dis 2009;68:635–41.



8 Gerloni V, Pontikaki I, Gattinara M, Fantini F. Focus on adverse events of tumour necrosis factor a blockade in juvenile idiopathic arthritis in an open monocentric long-term prospective study of 163 patients. Ann Rheum Dis 2008;67:1145–52. 9 Tynja¨la¨ P, Va¨ha¨salo P, Honkanen V, Lahdenne P. Drug survival of the first and second course of anti-TNF agents in juvenile idiopathic arthritis 2009. Ann Rheum Dis 2009; 68:552–7. 10 Quartier P, Taupin P, Bourdeaut F et al. Efficacy of etanercept for the treatment of juvenile idiopathic arthritis according to the onset type. Arthritis Rheum 2003;48: 1093–101. 11 Petty RE, Southwood TR, Manners P et al. for the International League of Associations for Rheumatology. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol 2004;3:390–2. 12 Guidance on the use of etanercept for the treatment of juvenile idiopathic arthritis. Technology Appraisal No. 35 National Institute for Clinical Excellence. 1–24 March 2002. www.nice.org.uk (17 June 2009, date last accessed). 13 Giannini EH, Ruperto N, Ravelli A, Lovell DJ, Felson DT, Martini A. Preliminary definition of Improvement in juvenile arthritis. Arthritis Rheum 1997;40:1202–9. 14 Clinical safety data management: definitions and standards for expedited reporting. CPMP/ICH/377/95. www. emea.eu.int (25 January 2010, date last accessed). 15 Lovell DJ, Ruperto N, Goodman S et al. for the Pediatric Rheumatology Collaborative Study Group and Pediatric Rheumatology International Trials Organisation. Adalimumab with or without methotrexate in juvenile rheumatoid arthritis. N Engl J Med 2008;359:810–20. 16 Ruperto N, Lovell DJ, Quartier P et al. for the Paediatric Rheumatology International Trials Organization and Pediatric Rheumatology Collaborative Study Group. Abatacept in children with juvenile idiopathic arthritis: a randomised, double-blind, placebo-controlled withdrawal trial. Lancet 2008;372:383–91. 17 Ruperto N, Lovell DJ, Cuttica R et al. for the Paediatric Rheumatology International Trials Organisation and Pediatric Rheumatology Collaborative Study Group. A randomized, placebo-controlled trial of infliximab plus methotrexate for the treatment of polyarticular-course juvenile rheumatoid arthritis. Arthritis Rheum 2007;56: 3096–106. 18 Tauber T, Daniel D, Barash J, Turetz J, Morad Y. Optic neuritis associated with etanercept therapy in two patients with extended oligoarticular juvenile idiopathic arthritis. Rheumatology 2005;44:405.

21 Bongartz T, Sutton AJ, Sweeting MJ et al. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. JAMA 2006;295:2275–85. 22 Dekker L, Armbrust W, Rademaker CM, Prakken B, Kuis W, Wulffraat NM. Safety of anti-TNF alpha therapy in children with juvenile idiopathic arthritis. Clin Exp Rheumatol 2004;22:252–8. 23 Horneff G, De Bock F, Foeldvari I et al. Safety and efficacy of combination of etanercept and methotrexate compared to treatment with etanercept only in patients with juvenile idiopathic arthritis (JIA). Preliminary data from the German JIA registry. Ann Rheum Dis 2009;68: 519–25. 24 Hyrich KL, Symmons DPM, Watson KD, Silman AJ on behalf of the British Society for Rheumatology Biologics Register. Comparison of the response to infliximab or etanercept monotherapy with the response to cotherapy with methotrexate or another disease-modifying antirheumatic drug in patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Arthritis Rheum 2006;54:1786–94. 25 Saurenmann RK, Levin AV, Feldman BM, Laxer RM, Schneider R, Silverman ED. Risk of new-onset uveitis in patients with Juvenile Idiopathic arthritis treated with anti-TNF Pediatr 2006;149:833–6. 26 Saurenmann RK, Levin AV, Rose JB et al. Tumour necrosis factor a inhibitors in the treatment of childhood uveitis. Rheumatology 2006;45:982–9. 27 Schmeling H, Horneff G. Etanercept and uveitis in patients with juvenile idiopathic arthritis. Rheumatology 2005;44: 1008–11. 28 Smith JA, Thompson DJ, Whitcup SM et al. A randomized, placebo-controlled, double-masked clinical trial of etanercept for the treatment of uveitis associated with juvenile idiopathic arthritis. Arthritis Rheum 2005;53: 18–23. 29 Reiff A, Takei S, Sadeghi S et al. Etanercept therapy in children with treatment-resistant uveitis. Arthritis Rheum 2001;44:1411–5. 30 Ruemmele FM, Prieur AM, Talbotec C et al. Development of Crohn disease during anti-TNF-alpha therapy in a child with juvenile idiopathic arthritis. J Pediatr Gastroenterol Nutr 2004;39:203–6. 31 Wiegering V, Morbach H, Dick A, Girschick HJ. Crohn’s disease during etanercept therapy in juvenile idiopathic arthritis: a case report and review of the literature. Rheumatol Int 2010;30:801–4.

19 Tauber T, Turetz J, Barash J, Avni I, Morad I. Optic neuritis associated with etanercept therapy for juvenile arthritis. J AAPOS 2006;10:26–9.

32 Kanakoudi-Tsakalidou F, Tzimouli V, Pratsidou-Gertsi P, Chronopoulou E, Trachana M. The significance of persistent newly developed autoantibodies in JIA patients under long-term anti-TNF treatment. Cytokine 2008;42: 293–7.

20 Armbrust W, Kamphuis SS, Wolfs TW et al. Tuberculosis in a nine-year-old girl treated with infliximab for systemic juvenile idiopathic arthritis. Rheumatology 2004;43:527–9.

33 Kimura Y, Pinho P, Walco G et al. Etanercept treatment in patients with refractory systemic onset juvenile rheumatoid arthritis. J Rheumatol 2005;32:935–42.


195