JIA-LANCET

Series 2138 www.thelancet.com Vol 377 June 18, 2011 Lancet 2011; 377: 2138–49 See Comment page 2067 This is the third in a Series of three papers about arthritis Centre for Molecular and Cellular Intervention, Department of Paediatrics, University Medical Centre Utrecht, Netherlands (Prof B Prakken MD); Translational Research Laboratory, Infl ammatory and Infectious Diseases Center, Sanford-Burnham Medical Research Institute, La Jolla, CA, USA (Prof S Albani MD); Department of Paediatrics, University of Genova and Pediatria II, Institute G Gaslini, Genova, Italy (Prof A Martini MD) Correspondence to: Prof Alberto Martini, Department of Paediatrics, University of Genova and Pediatria II, Institute G Gaslini, Largo G Gaslini 5, 16147 Genova, Italy albertomartini@ospedale- gaslini.ge.it Arthritis 3 Juvenile idiopathic arthritis Berent Prakken, Salvatore Albani, Alberto Martini Juvenile idiopathic arthritis is a heterogeneous group of diseases characterised by arthritis of unknown origin with onset before age of 16 years. Pivotal studies in the past 5 years have led to substantial progress in various areas, ranging from disease classifi cation to new treatments. Gene expression profi ling studies have identifi ed diff erent immune mechanisms in distinct subtypes of the disease, and can help to redefi ne disease classifi cation criteria. Moreover, immunological studies have shown that systemic juvenile idiopathic arthritis is an acquired autoinfl ammatory disease, and have led to successful studies of both interleukin-1 and interleukin-6 blockade. In other forms of the disease, synovial infl ammation is the consequence of a disturbed balance between proinfl ammatory eff ector cells (such as T-helper-17 cells), and anti-infl ammatory regulatory cells (such as FOXP3-positive regulatory T cells). Moreover, specifi c soluble biomarkers (S100 proteins) can guide individual treatment. Altogether these new developments in genetics, immunology, and imaging are instrumental to better defi ne, classify, and treat patients with juvenile idiopathic arthritis. Introduction Juvenile idiopathic arthritis is not a single disease, but a term that encompasses all forms of arthritis that begin before a patient is aged 16 years that persist for more than 6 weeks and are of unknown origin.1,2 It is the most common childhood chronic rheumatic disease and causes much disability. In high-income countries it has a yearly incidence of 2–20 cases per 100 000 population and a prevalence of 16–150 cases per 100 000 population.1 In this Seminar we focus on developments in the understanding of pathogenesis and in the diagnosis and treatment, and discuss how translational research and new imaging modalities and biomarkers are expected to improve diagnostic and treatment options. Clinical manifestation and classifi cation Disorders described by the term juvenile idiopathic arthritis have been grouped on the basis of clinical and laboratory features to try and identify homogeneous, mutually exclusives categories.3 Clinical and laboratory fi ndings have improved the understanding of the diff erent forms of chronic childhood arthritis.4–6 Although some categories identify defi nite disease entities, others still include heterogeneous disorders.7 Well-characterised categories Systemic juvenile idiopathic arthritis is characterised by prominent systemic features, such as fever, rash, and serositis, and is much like adult-onset Still’s disease.1 Pronounced activation of a patient’s innate immune system and the absence of any consistent association with autoantibodies or human leucocyte antigen (HLA) have led to the hypothesis that this systemic form of disease is a polygenic autoinfl ammatory syndrome.8 Findings from previous studies suggesting a major pathogenic role for interleukin-69 in the disease have been substantiated by the reported eff ectiveness of treatment with tocilizumab, an anti-interleukin 6 receptor antibody.10,11 Moreover, the fi nding that anti-interleukin 1 treatment can also be eff ective12 has led to the subsequent delineation of two subpopulations of this form of disease:13 one with a pronounced, complete response to interleukin-1 blockade (much the same as seen in cryopyrin-associated autoinfl ammatory syndromes) and another that is resistant to treatment or has an intermediate response. These two subpopulations do not diff er in interleukin-1 in vitro production or in serum cytokine concentrations, but only in the number of joints aff ected and in neutrophil counts; patients with fewer joints aff ected or with a higher neutrophil count have an increased probability of responding to anti-interleukin-1 treatment. Thus, systemic juvenile idiopathic arthritis can be stratifi ed into at least two subgroups on the basis of responsiveness to inhibition of—and therefore possible pathogenic relevance of—inteleukin-1. Once the fi ndings from a phase 3 trial of canakinumab (an anti- interleukin-1 antibody; NCT00889863) and a second phase 3 trial of tocilizumab (NCT00642460) are available, the pattern of response might provide new clinical or laboratory biomarkers useful to further understand the Search strategy and selection criteria We searched PubMed for papers with the search terms “juvenile arthritis”, “systemic”, ”polyarticular”, “oligoarticular”, “psoriasis”, “spondylarthropathy”, “therapy”, “pathogenesis”, “immunology”, “genetics”, “cytokines”, and “T cells”. We gave preference to papers published between 2005 and 2010, without excluding key references from earlier years. We also gave preference to original studies published in peer-reviewed journals, but also searched for, and if appropriate included, abstracts from the main paediatric rheumatology conferences during the past 3 years. We aimed to include the most recent publications while also referring to the fi rst original publication on a given subject. 123 下划线 123 下划线 123 波浪线 123 高亮 123 高亮 123 高亮 123 波浪线 Series www.thelancet.com Vol 377 June 18, 2011 2139 heterogeneity between these two subgroups and the reciprocal roles of interleukin-1 and interleukin-6 in disease pathogenesis. Rheumatoid factor positive polyarthritis, a small subcategory of juvenile idiopathic arthritis (aff ecting 5% of patients), is thought to be much like adult rheumatoid factor (RF)-positive rheumatoid arthritis; indeed, it is the only form of juvenile idiopathic arthritis with positive antibodies to cyclic citrullinated peptides.14 Major diff erences with the adult form of disease are in the eff ect that the disease can have on a growing skeleton, leading to either general growth retardation or accelerated growth of an aff ected joint. Enthesitis-related arthritis is a form of undiff erentiated spondyloarthropathy.15 Most patients are HLA-B27 positive, and, in about 30–40% of patients, the disease can progress to aff ect sacroiliac joints. Although the category oligoarthritis, as a whole, is probably heterogeneous, most patients—at least in developed countries—have a well-defi ned disease that is seen only in children.16 Arising more often in girls than in boys, oligoarthritis has early onset (before 6 years of age), has consistent HLA associations, and is characterised by asymmetric arthritis that aff ects mainly large joints. Patients have high concentrations of positive antinuclear antibodies (ANA) and a high risk of developing chronic iridocyclitis. Juvenile idiopathic arthritis classifi cation criteria3 distinguish two categories of oligoarthritis: persistent oligoarthritis, in which the disease aff ects four joints or fewer, and extended oligoarthritis, in which more than four joints are aff ected after the fi rst 6 months of disease. However, patients with either persistent or extended oligoarthritis who test positive for ANA have similar clinical characteristics (eg, age at onset, sex ratio, asymmetry of articular involvement, and frequency of iridocyclitis), which strongly suggests that these two categories of oligoarthritis are the same disease, diff ering only in severity.17,18 Less well-characterised categories Rheumatoid factor-negative polyarthritis is a hetero- geneous category of juvenile idiopathic arthritis. At least two subsets can be identifi ed: one that is similar to adult- onset RF-negative rheumatoid arthritis, characterised by a symmetric synovitis of large and small joints, onset at school age, and the absence of ANA expression, and another that resembles oligoarthritis, apart from the number of joints aff ected in the fi rst 6 months of disease. Similarities between this second subset and early-onset oligoarthritis led to the hypothesis that they are the same disease, with a more rapid spread of arthritis in the second subset than in early-onset arthritis.16 This view has been lent support by the fi nding that ANA-positive oligoarthritis share the same features with ANA-positive RF-negative polyarthritis, but not with ANA-negative RF-negative polyarthritis or with ANA-negative oligoarthritis.17 The view is also lent support by the fact that ANA-positive RF-negative polyarthritis is seldom seen in countries in which ANA-positive oligoarthritis is rare.16 If psoriatic arthritis is defi ned according to the presence of arthritis and psoriasis or some psoriatic features (as it is in the Vancouver criteria19), two disease entities exist:16,20,21 one in the enthesitis-related arthritis category, which is therefore, like adult psoriatic arthritis, a form of spondyloarthropathy, and a second that is very similar to ANA-positive oligoarthritis with only small diff erences such as it aff ects small joints more often than large joints, a feature that could be attributable to psoriatic diathesis in the ANA-positive oligoarthritis phenotype. The association of psoriasis with arthritis seems to lead to the identifi cation of two subsets of patients—one with disease that is similar to adult psoriatic arthritis and another with disease that has only minor diff erences with ANA-positive oligoarthritis.16 Indeed, most patients who meet the present classifi cation criteria for psoriatic arthritis, in which patients with enthesitis are by defi nition excluded, have features of ANA-positive oligoarthritis. Perspectives for a new classifi cation To improve our understanding of the cause and development of the various forms of childhood chronic arthritis and fi nd more suitable treatments, the identifi cation of categories that, at least from a clinical point of view, seem as homogeneous as possible is essential to enable immunological, gene expression, and genome-wide association studies. If more homogeneous groups within juvenile idiopathic arthritis are to be identifi ed, which seems likely in view of the heterogeneity within the present subcategories, then some classifi cation criteria need to be reconsidered. In 2003, we suggested16 that the number of joints aff ected and the presence of psoriasis are not suitable criteria with which to identify homogeneous disease entities, and that children with clinical features that are strongly suggestive of a common cause (eg, asymmetric arthritis, early onset, sex ratio, ANA positivity, high risk for iridocyclitis) are wrongly classifi ed into three diff erent disease categories— oligoarthritis, RF-negative poly arthritis, and psoriatic arthritis. We postulated that the grouping of patients into these three categories according to criteria (alone or in combination) such as ANA positivity, age at disease onset, or pace (asymmetrical or symmetrical) at which the disease aff ects joints could lead to the defi nition of more homogeneous categories. The analysis of gene expression profi les has confi rmed the heterogeneity of polyarticular juvenile idiopathic arthritis.22 Moreover, Barnes and colleagues23 recorded a B-cell signature that characterises patients with early- onset arthritis independently from the number of joints aff ected. Their study accords with previous fi ndings showing that plasma cell infi ltration of the synovium was more common in the early phase of joint infl ammation but not related to disease activity or severity.24 Additionally, 123 高亮 123 波浪线 123 高亮 123 波浪线 123 高亮 123 高亮 Series 2140 www.thelancet.com Vol 377 June 18, 2011 high-resolution HLA class I and class II typing has shown similarities between early-onset polyarticular and oligoarticular forms of disease.25 All these fi ndings corroborate the suggestion that ANA-positive, early-onset arthritis is a homogeneous entity that is classifi ed into diff erent disease categories because of diff erences in the spread of arthritis or the association with psoriasis or psoriatic features. Ultrasonography has shown much discrepancy between imaging and clinical examination in the assessment of the number of aff ected joints,26 making a classifi cation on the basis of the number of joints aff ected even more complicated. However, ultrasonography allows better diff erentiation between tendon and articular involvement, thus providing information that could also be of relevance for classifi cation purposes.27 Because in children, as in adults, several diff erent diseases exist that all cause chronic arthritis, the terms juvenile idiopathic arthritis and juvenile idiopathic arthritis onset-forms will probably become outdated as more is learnt about each disease. These terms wrongly suggest that juvenile idiopathic arthritis is a single disease (as was thought many years ago) and that the various onset-forms (or categories) are only pheno- typic variants. Cause and pathogenesis One of the most intriguing questions in the study of human autoimmune diseases is what determines the phenotype and organ specifi city of a disease. In juvenile idiopathic arthritis, for example, the occurrence of uveitis is related to explicit risk factors, such as age at disease onset, sex, the presence of ANA auto-antibodies, and the subtype of juvenile idiopathic arthritis.28–30 Despite these associations, the immune pathogenesis underlying the link between arthritis and uveitis is unknown. However, studies done in the past 5 years have provided information about the immune pathogenesis of juvenile idiopathic arthritis, confi rming, among other things, that the systemic form is a diff erent disease from other types of juvenile idiopathic arthritis, with a distinct immune pathogenesis, and should be treated as would an acquired autoinfl ammatory disease. Genetic susceptibility Despite the heterogeneity of juvenile idiopathic arthritis, a specifi c genetic overlap might exist between subtypes since they share joint infl ammation as the most prominent disease feature.31 The concordance rate and similarity in disease phenotype between monozygotic twins (20–40%) also suggests a strong genetic component of the disease.32,33 Indeed, fi ndings from several studies suggest that specifi c genetic susceptibility genes can be identifi ed, which can be broadly divided into two groups—HLA genes and non HLA-related genes. The non HLA-related genes include cytokines and other immune genes.34 The association of juvenile idiopathic arthritis with both HLA class I (HLA A-2, and HLA B27) and HLA class II (HLADRB1 and HLA DP) alleles has been reported and substantiated in various studies, underscoring the supposed importance of the role of T cells in the disease course.35 Many non-HLA candidate genes have been studied for associations with juvenile idiopathic arthritis, including various cytokine genes, but only a few candidate genes including PTPN22, MIF, SLC11A6, WISP3, and tumour necrosis factor (TNF)α could be independently confi rmed.36,37 Instead of searching for specifi c candidate genes, doing genome- wide association studies are an attractive alternative.38 This approach has led to the successful identifi cation of new genes and immune pathways in other autoimmune diseases. Because the eff ects of individual genes in a complex autoimmune disease such as juvenile idiopathic arthritis are probably very small, international collaborations will be necessary for suffi ciently powered studies. As discussed in the Treatment section of this Seminar, such international collaborations are indeed actively pursued in paediatric rheumatology. Gene expression profi ling In addition to searching for susceptibility genes, gene expression profi ling can identify any genes for which expression patterns change during the course of disease, providing a disease signature that can lead to the identifi cation of unique biomarker patterns. Although this approach has disadvantages (such as the high costs, the complexity of the composition of the DNA microarrays, and the analysis of the results), its potential is underscored by various studies in patients with juvenile idiopathic arthritis—studies of patients with diff erent forms of disease showed a consistent, though not fully elucidated, association with interleukin-10-regulated genes in all forms of the disease.39–41 Also, peripheral blood mononuclear cells from patients with recent-onset disease have a diff erent gene expression profi le dependent on their subtype.41 In another study, researchers compared early samples of either peripheral blood mononuclear cells or synovial fl uid mononuclear cells from patients who developed either persistent or widespread oligoarticular disease and identifi ed patterns that were diff erent between both groups.42 However, the most striking fi nding was an increased CD4 to CD8 ratio in patients with remitting disease, underscoring the important part played by T cells in disease course.42 Also, as discussed above, gene expression profi les in peripheral blood mononuclear cells of patients with recent-onset juvenile idiopathic arthritis identifi ed expression patterns that diff ered with age of onset but not with the numbers of joints aff ected, suggesting diff erent biological mechanisms in patients with either early-onset and late-onset disease.22,23 Gene expression patterns in systemic juvenile idiopathic arthritis are diff erent from those of other subtypes, and include upregulation of genes related to Series www.thelancet.com Vol 377 June 18, 2011 2141 innate immunity and complement systems, and of a group of mostly haemopoietic genes.43 The next challenge will be to relate these molecular profi les to specifi c immune pathways and the clinical manifestations to individual pathways. Thus gene expression profi ling can provide a basis for a novel molecular method of classifi cation of patients at disease onset, which could help to better predict treatment response and thus guide individual treatment regimens.4,44 Environmental triggers In a healthy immune system, eff ector and regulator mechanisms are kept in balance, assuring a tailor-made response that adequately protects against an invading pathogen while preventing unwarranted damage to an individual and preserving immune tolerance. To achieve this balance, the innate and adaptive immune systems closely interact. Much the same as with most human autoimmune diseases, the cause of juvenile idiopathic arthritis is assumed to be multifactorial. A genetically susceptible individual might develop a deleterious and uncontrolled response towards a self-antigen on exposure to an unknown environmental trigger.32 This response causes a self-perpetuating loop of activation of both innate and adaptive immunity that causes tissue damage. In juvenile idiopathic arthritis, infections and vaccinations have been suggested as two candidate triggers, but neither has been confi rmed as a trigger because of a scarcity of proper controlled, prospective studies.4,32,45 A prospective study did not show a relation between vaccination with meningoc