Quite well characterised JIA categories

Systemic JIA

Systemic JIA (sJIA) is a disease clearly distinguished from all the other forms of JIA and very similar to adult-onset Still's disease. It is characterised by prominent systemic features, such as high spiking fever, and an important elevation of acute phase reactants. There is a marked activation of the innate immune system,4 witnessed also by the frequent occurrence (in about 5–8% of patients) of a life-threatening complication known as macrophage activation syndrome—a form of haemophagocytic lymphohistiocytosis. The reason for the high rate of occurrence of macrophage activation syndrome in sJIA remains unknown.5 The hypothesis that interleukin 6 (IL-6) has a major role in the pathogenesis of sJIA6 has been confirmed by the efficacy of tocilizumab, an anti-IL-6 receptor antibody.7 Anti-IL-1 treatment has also proved effective, and gene expression analysis in active disease has shown an IL-1 signature.8 Moreover, treatment with anakinra has distinguished two subsets of patients9: one that has a dramatic response similar to that observed in cryopyrin-associated autoinflammatory syndromes and the other that is resistant or has an intermediate response. The systemic features, the marked activation of the innate immune system, the lack of any consistent association with autoantibodies or human leucocyte antigen (HLA) antigens, and the response to IL-1 inhibition has led to the categorisation of at least some forms of sJIA as autoinflammatory diseases.

Enthesitis-related arthritis

Enthesitis-related arthritis is a form of undifferentiated spondyloarthropathy. Most patients are HLA-B27 positive and for some, involvement of the sacroiliac joints develops during the disease course.10

Rheumatoid factor-positive polyarthritis

Rheumatoid factor (RF)-positive polyarthritis is the same disease as adult RF-positive rheumatoid arthritis (RA). Indeed, it is also the only form of JIA with positive antibodies to cyclic citrullinated peptides.11

Oligoarthritis

The large majority of patients with oligoarthritis belong to a quite well defined disease which is typical of children and is not seen in adults. It is characterised by an asymmetric arthritis affecting mainly large joints, an early onset (before 6 years of age), a female predilection, a high frequency of positive antinuclear antibodies (ANA), a high risk for developing chronic iridocyclitis and consistent HLA associations. The International League of Associations for Rheumatology (ILAR) classification distinguishes two categories of oligoarthritis based on the number of joints that are involved after the first 6 months of disease: persistent oligoarthritis, in which the disease remains confined to four or less joints, and extended oligoarthritis, in which arthritis extend to more than four joints (which may occur in up to 50% of patients). However, it has been shown that ANA-positive patients belonging to these two categories share the same characteristics, strongly suggesting that they represent the same disease.12

Less well-characterised JIA categories

RF-negative polyarthritis

RF-negative polyarthritis is a heterogeneous JIA category. At least two distinct subsets can be identified: (a) a form that is similar to adult-onset RF-negative RA and is characterised by symmetric synovitis of large and small joints, onset in school age and negative ANA; (b) a form that resembles oligoarthritis, except for the number of joints affected in the first 6 months of disease. These similarities led to the hypothesis that this second subset of RF-negative polyarticular JIA and early-onset oligoarthritis are the same disease, the former representing a rapid arthritis spread in the latter.3 This view has been confirmed by the demonstration that ANA-positive oligoarthritis (either persistent or extended) shares the same features (asymmetric arthritis, early age at onset, female predominance, ANA positivity, elevated incidence of chronic iridocyclitis) as ANA-positive, RF-negative polyarthritis, but not the same as ANA-negative RF-negative polyarthritis or ANA-negative oligoarthritis.12 The concept is also supported by the observation that in those countries in which ANA-positive oligoarthritis is rare, ANA-positive, RF-negative polyarthritis is also seldom seen.3 The heterogeneity of RF-polyarticular JIA has recently been confirmed also by gene expression studies.13

Psoriatic arthritis

Psoriatic arthritis also does not appear to represent a defined entity.3 If psoriatic arthritis is defined according to the presence of arthritis and psoriasis or some psoriatic features (Vancouver criteria) two populations of patients are identified: (a) one that belong to the enthesitis-related arthritis category and represent therefore, like adult psoriatic arthritis, a form of spondyloarthropathy; (b) a second form that is very similar to ANA-positive oligoarthritis with only small differences such as a more frequent involvement of small joints, a feature that may well reflect the influence of the psoriatic diathesis on the ANA-positive oligoarthritis phenotype. Consistently, most patients who meet the ILAR criteria for psoriatic arthritis, in which patients with enthesitis are by definition excluded, have the features of ANA-positive oligoarthritis. So it appears that the association of psoriasis with arthritis leads to the identification of two different subsets of patients—one that is similar to adult psoriatic arthritis and the other that is overlapping, with only minor differences, with ANA-positive oligoarthritis.3 14

Undifferentiated arthritis

Undifferentiated arthritis is by definition heterogeneous since it includes patients who do not fulfil inclusion criteria for any category or fulfil the criteria for more than one category.

Treatment perspectives

It was only slightly more than a decade ago that the introduction of etanercept for the treatment of children with JIA15 began to dramatically change both the observed responses to treatment and the expectations of treatments.16 During this same era the transition from the use of juvenile RA categories to the use of JIA categories in the determination of eligibility in clinical trials occurred. Lack of clarity about the efficacy of various biological treatments in subsets of patients with JIA is a consequence of both this transition in the use of classification criteria and the biological heterogeneity of both the juvenile RA and JIA subgroups as described above.

The performance of these clinical trials has benefited greatly from several factors. Key non-pharmaceutical supported studies have been performed by the Pediatric Rheumatology InterNational Trials Organization (PRINTO) and the Pediatric Rheumatology Collaborative Study Group (PRCSG) which have greatly enhanced our ability to study JIA. For example, the definition of the JIA core set of variables, definitions of response, definition of ‘inactive disease’ and clinical remission states are critically important to the trials that are being performed currently. The identification in a large randomised clinical trial of the maximally effective dose of methotrexate in polyarticular forms of JIA provided a firm understanding of the most appropriate prebiological treatment approach for many children with JIA.17 In addition, a validated core set of clinical measures as well as validated definitions for treatment response—the Pediatric 30 definition of response18 and disease flare15 19—have served to greatly improve both the scientific validity and efficiency of the performance of these trials.

Furthermore, the trial of etanercept in JIA introduced to paediatric rheumatology the use of the double-blind, randomised withdrawal study design.19 This treatment design minimises both the number of children required to demonstrate a significant difference in clinical effect between an active agent and placebo and also the time for which children are required to remain on placebo. Both of these aspects are relevant when either the efficacy or safety of a new treatment is uncertain (clinical equipoise). In addition, this trial design allows all study subjects to be treated with the new active agent and the status of the individual study subject is the determinant of the duration that a particular study subject remains in the double-blind, placebo-controlled part of the study. This study design has proved to be very well accepted by study participants and their families (user friendly). It has also proved to be ethically acceptable to investigators and ethics review committees/institutional review boards since it minimises the time that a subject experiences active arthritis induced by the study design. This study design has been sufficient proof to regulatory agencies—both the Food and Drug Administration (FDA) and the European Medicines Agency (EMA)—for them to approve etanercept,15 adalimumab20 and abatacept21 for use in polyarticular forms of JIA. Moreover, PRINTO as well as the Childhood Arthritis and Rheumatology Research Alliance are actively engaged in non-industry-supported clinical trials.

Of critical importance to the successful completion of these studies in JIA has been the existence of two international research networks of paediatric rheumatology centres—PRCSG and PRINTO—that have worked in a highly integrated and synergistic fashion on design and performance of preliminary studies (phase II), pivotal clinical trials (phase III) and clinical-effectiveness and safety studies (pen label extension and phase IV). With the exception of etanercept, PRCSG and PRINTO have collaborated on the testing of all the biological agents used in children with JIA to date.21,–,23

The Pediatric 30 level of response is seen in over 80% of the children with polyarticular forms of JIA enrolled in the clinical trials.15 20,–,25 To provide for more demanding levels of response, Pediatric 50, 70, 90 and 100 levels of response have been assessed in the clinical trials.15 20,–,25 More recently, a validated preliminary definition of ‘clinically inactive disease’26 27 has been used as the primary outcome measure in a randomised, double-blind trial in children with early JIA—the National Institutes of Health funded TREAT trial (Wallace PI). The trial is still continuing but it represents the first trial to use inactive disease as the outcome. Increasingly, in clinical studies we need to focus on higher levels of response and in the near future, investigator studies are needed to expand the TREAT trial to see if induction of drug-free remission is achievable.

Trials of children with sJIA with active systemic features have demonstrated the efficacy of IL-6 blockade in a phase III study in Japan.7 International phase III studies are continuing for both IL-1 and IL-6 blockade.

Safety has been assessed in phase III,7 20,–,23 open-label extension18 24 25 28 and registry studies29,–,32 of biological agents in children with JIA. To date, no unique safety signals have been recognised in children that have not already been identified in adults, with the exception of a possibly increased risk of lymphoma and other types of cancer in children treated with one of the anti-tumour necrosis factor (TNF) agents. This observation is based on spontaneous adverse event reports (SERS) in the FDA surveillance system. However, the SERS system has major limitations, including uncertainty of the completeness of the adverse event reporting (numerator data) and lack of an internal estimate of the at-risk population (denominator data). Moreover, the rate of malignancies in children with JIA not receiving biological treatments is not well established. The long-term safety of biological agents and of the other drugs used in JIA needs to be assessed with the use of large international registries.

Future perspectives

If more homogeneous groups within JIA have to be identified, then some classification criteria have to be reconsidered. In 2003 we suggested3 that the number of joints affected and the presence of psoriasis do not represent suitable classification parameters to identify homogeneous disease entities and that children with the same cluster of features that strongly suggest a common background (asymmetric arthritis, early onset, female predominance, ANA positivity, high risk for iridocyclitis, definite HLA associations) are currently classified into different JIA categories (oligoarticular, polyarticular RF negative and psoriatic). Moreover, ultrasonography has shown a frequent discrepancy between imaging and clinical examination in evaluation of the number of affected joints,33 making a classification based on the number of joints involved even more complicated. We suggested3 that grouping patients according to other criteria (alone or in combination) such as ANA positivity, age at onset or pace (asymmetrical or symmetrical) of joint involvement might lead to the definition of more specific categories. Indeed, as mentioned above, ANA positivity has been shown to identify a homogeneous subset of patients1 12 and, more recently, it has been shown that a B-cell gene expression signature can distinguish between early-onset and late-onset JIA regardless of the number of joints affected.34 So, further refinements in the classification criteria are needed in order to identify more homogeneous entities suitable for immunological, gene expression and genome-wide association studies.35 36

The potential of imaging has still to be fully exploited in childhood arthritis. Indeed, methods to evaluate x-ray findings have progressed,37 MRI has shown the potential for identifying bone damage with high sensitivity38 and ultrasonography has demonstrated its utility in assessing the number of active joints and in differentiating tendon from articular involvement.33 In the future, molecular imaging will probably be able to identify bone and cartilage involvement well before morphological damage became apparent. Serum biomarkers (such as S100 proteins) are also showing promise in identifying subclinical disease and predicting relapse.39 40

Progress in understanding the pathogenesis of RF+ RA and the spondyloarthropathies will also shed light on their childhood equivalents.

The efficacy in the various JIA categories of TNF, IL-6 and IL-1 blockade has provided excellent therapeutic tools and also helped in unravelling disease pathogenesis. So it can be expected that the future availability of new drugs with other specific targets will continue to foster this process of ‘reverse translation’ (from the bed to the bench side).

Systemic arthritis will represent a precious opportunity to better understand the role of innate immunity alterations in the pathogenesis of chronic inflammation as well as the factors that in that context predispose to the development of persistent and aggressive synovitis.

The fact that an homogeneous disease such as ANA-positive oligoarthritis can have a more benign (persistent oligo) or a more aggressive (extended oligo) course offers a useful model for differentiating factors related to disease aetiology from those that influence disease progression. Interesting results have been indeed achieved for regulatory T cells,41 42 their relationship with Th17 cells43 and the role of heat shock proteins in modulating the immune response.44 45

Terms such as JIA and onset forms will probably be abandoned in the future since they suggest the misleading concept that the various forms of chronic childhood arthritis represent different subsets of a single disease rather than completely different diseases.

Therapeutic trials have already established the efficacy of anti-TNF agents in children with the polyarticular forms of JIA. Future studies of additional anti-TNF agents that have established efficacy and safety in adults with RA should only need to perform dosing and safety studies in patients with polyarticular forms of JIA. As additional classes of treatment establish efficacy in children with JIA (eg, IL-1 blockade in sJIA), then fewer efficacy trials will need to be performed.

Current efforts to establish very large multinational longitudinal cohorts of patients with JIA will provide the only realistic approach to assessing the true relative safety of all the agents used to treat children with JIA, including the various biological treatments, and to assessing the risk for rare adverse events. In addition, this effort will be greatly enhanced by the development of banks of biological specimens obtained from patients with JIA that have been collected using validated and standardised techniques for specimen collection, preparation and storage, such as the Canadian International Partnership Initiative.

The past decade has brought dramatic advances in the understanding of JIA categories, pathobiology and treatments but much remains to be done.