What makes severe JIA severe?

As mentioned in the introduction, we will mainly focus on systemic JIA (sJIA) and severe extended oligoarticular or polyarticular JIA. Notwithstanding this focus, it should be stressed that also oligoarticular JIA can be a serious disease burden for children, as the loss of function in even a single joint may lead to serious disability. Moreover, especially oligoarticular JIA patients are at a risk of developing uveitis as extra-articular complication or disease manifestation, which can threaten the vision in a significant percentage of children .

The most important reason for the focus on polyarticular and sJIA is pragmatic as, in general, the disease burden in JIA increases with the number of affected joints and with more severe systemic inflammation (as in sJIA). These patients have the highest risk of irreversible damage and long-term morbidity, and, consequently, they are most at risk of a lower quality of life. As such, these are the children with the greatest need for improved management.

New challenges in a changing landscape

The increased therapeutic possibilities of the past decade, resulting in improvement of arthritis in most children, change the way we define and classify severe JIA. A new description should thus not only consider active disease and joint damage but should take the chronicity of undergone treatment into account as well. For example, even as nowadays most patients can achieve clinically inactive disease, many of these patients will experience a relapse once treatment is tapered or stopped. This suggests that we achieve a state of disease suppression to a level that is clinically not detectable rather than a real cure. Even without the occurrence of irreversible joint damage through improved treatment, we do not know what the effects of continuous treatment with MTX or biologicals for 5, 10 or more years will be on the health of these children. Does it affect, for example, fertility? Or does it have an impact on the risks of cardiovascular events? Or even on the risk of developing a malignancy? In other words, we do not know whether and if so how the combination of disease and long term-therapy influences their healthy ageing.

The amazing progress of the past decade thus is not sufficient. Now, it is vital to redefine and reset the goals of treatment. Whereas conventional treatment of JIA aimed to suppress clinical overt disease activity, the new goal will be to achieve not only clinical but also immunological inactive disease. This should, in the end, open the door towards a disease remission without treatment, in other words a cure.

To realize this, crucial progress is needed on several areas ( Table 1 ). First of all, it is crucial to reach a better understanding of the disease pathogenesis. JIA is a complex disease in which environment and genetic susceptibility in concert lead to disease. Although major steps have been made in the past years, the question of what determines disease in an individual patient still remains a puzzle. Second, we need to have a better understanding of the mechanisms of action of the various drugs used in JIA. It is surprising and disappointing that the exact working mechanism of commonly used drugs such as MTX and anti-TNF therapy on the immune system is still largely unknown. Third, biomarkers are needed that can help to profile individual patients to predict their prognosis, response to treatment and risks on side effects. Lastly, true long-term multicenter studies and medication registries are necessary to determine the long-term consequences of the disease and its treatment.

Altogether, this should help to define personalized strategies for tapering or intensifying treatment as well as help to develop more specific maintenance therapies. This could involve combining therapeutic modalities early in the disease course to modulate immune responses with the goal to change the biology of the disease. This would be followed by a step down in therapy when clinically and biochemical inactive disease criteria are met. Therefore, the development of therapeutic maintenance modalities that are theoretically less risk full will be valuable, for example, antigen-specific immune therapies directed at strengthening and not just suppressing immune responses. To date, these types of approaches are no longer science fiction, as the examples discussed below will prove that improved understanding can go hand in hand with clinical benefits.

sJIA versus other JIA subtypes

The first part of this review is committed to sJIA. This intriguing disease, with impressive systemic inflammation besides chronic arthritis, is still classified under the umbrella of JIA. However, evidence is accumulating that at least in the initial and early phases of sJIA, the mechanisms of disease significantly differ from mechanisms of disease in non-sJIA. Most immune aberrations in sJIA originate from the innate immune system, resembling auto-inflammatory diseases in many aspects. As we will discuss below, an increased understanding of sJIA has led to concrete changes in its treatment.

sJIA is a complex, polygenic auto-inflammatory disease

A case can be made for the classification of sJIA as a polygenic auto-inflammatory disease. Arguments for this are summarized in Table 2 . First of all, in the initiation and early phase of this disease, there is no evidence for auto-reactive T cells or auto-reactive antibodies . Moreover, the lack of human leucocyte antigen (HLA) associations with sJIA also argues against a role for autoreactive T cells. Instead, especially polymorphisms in the putative promoter regions of several cytokines have been associated with sJIA . In addition, microarray studies using peripheral blood mononuclear cells of sJIA patients show prominent innate immune activity and a limited role for adaptive immunity .

A genome-wide association study for sJIA is still under way, but linkage studies as well as the (lack of any) inheritance patterns in patients with sJIA indicate that sJIA is a polygenic, not a monogenic disease.

Mechanistically, the prototypical auto-inflammatory disorders are characterized by activation of the interleukin (IL)-1β pathway . In fact, there is ample evidence that the IL-1 pathway is crucial in sJIA as well . The strong IL-1 signature in the pathophysiology of sJIA was first shown by Virginia Pascual in 2005 , and since then evidence that many features of sJIA are IL-1 mediated has strengthened .

On a cellular level, especially monocytes/macrophages and neutrophils are increased and activated in active sJIA , whereas natural killer (NK) cells are deficient in both numbers and function . In summary, there is enough evidence from both genetic and mechanistical studies to classify sJIA as an acquired auto-inflammatory disease rather than an autoimmune disease.

The inflammasome is activated in sJIA

One of the central features of many classical auto-inflammatory diseases is increased inflammasome activation. In cryopyrin-associated periodic syndromes (CAPS), over-activation of the NLRP3 (nucleotide-binding domain and leucine-rich repeat-containing protein 3) inflammasome is caused by gain-of-function mutations in the NLRP3 gene. Normally, the NLRP3 inflammasome has a function in sensing of various microbial products as well as endogenous danger signals (damage-associated molecular patters, or DAMPs), resulting in the activation of Caspase-I and thereby initiating IL-1β and IL-18 processing . In the classical or intrinsic inflammasomopathies (with mutations in NLRP3), especially IL-1β plays a central role in the clinical manifestations. However, it now has become clear that extrinsic inflammasomopathies (without mutations in NLRP3) also exist, sometimes with a complex genetic background. In this category, sJIA also fits, albeit as a complex, not a monogenic disorder .

The basis of the observed excess IL-1 activity and IL-18 levels is not exactly known. The group of Johannes Roth from the University of Muenster in Germany suggested S100A8/9 complexes, known to bind TLR4 and resulting in increased pro-IL-1β and pro-IL-18 production , are part of a positive feedback loop for overproduction of IL-1β . Although not demonstrated in monocytes of sJIA patients in a study by Gattorno et al. , indeed increased Caspase-I activation can be present in white blood cell lysates of active sJIA patients (Vastert, de Jager et al., submitted for publication). This could contribute to the increased levels of IL-18 by cleaving pro-IL-18 to IL-18, but it does not explain why IL-1β can hardly be picked up in plasma.

In vivo, pro-IL-1 and pro-IL-18 may also be cleaved by other pathways, such as proteases in inflamed joints . In any way, these observations point to the involvement of increased inflammasome activation in sJIA, contributing to its auto-inflammatory nature.

There is more to sJIA than IL-1 mediated inflammation

The picture that arises of sJIA from ours and other studies is that of an auto-inflammatory, IL-1-mediated disease and less that of a classical autoimmune arthritis. However, phenotypically and immunologically sJIA differs from the prototypical monogenic auto-inflammatory diseases as well. For one thing, without denying the pivotal role of IL-1β in the pathophysiology of sJIA, other (soluble) factors like cytokines and inflammatory proteins are important as well.

Another important contributor to the phenotype of sJIA is IL-6. Besides the association of increased levels of IL-6 with iron deficiency-related anaemia, growth failure and fever , the impressive response rates to therapeutic IL-6-blockade in clinical trials exemplify this . Besides IL-6, as mentioned above, more recently IL-18 attracts attention.

IL-18 as a biomarker of disease activity in sJIA

IL-18 is a member of the IL-1 cytokine family and shares similarities in genetic sequence, protein processing, receptor signalling and intracellular pathways with the IL-1β pathway . However, it functionally resembles IL-12 as well, especially relating to interferon gamma (IFN-γ) production and activation of NK cells and macrophages . Moreover, IL-18 is a potent activator of neutrophils, leading to migration, degranulation and cytokine/chemokine release .

A cross-sectional study of sJIA patients in 2007 showed highly elevated levels of IL-18 in active sJIA. Even sJIA patients in remission display elevated levels of IL-18, whereas IL-1β can hardly be picked up in peripheral plasma in active sJIA patients . Other reports also have suggested that IL-18 is a biomarker of disease activity in sJIA . Moreover, IL-18 in plasma could serve as a biomarker for response to therapy as well. In contrast to the cross-sectional study with persistent increased levels of IL-18 in remission on steroids, IL-18 normalizes in patients treated with rIL-1RA as first-line therapy and remains low after stopping rIL-1RA . This points to a potentially persistent change in the immune mechanisms underlying systemic inflammation in sJIA. Clearly, these findings will need confirmation, preferably in another controlled prospective cohort of sJIA patients.

IL-18 as a diagnostic tool in early sJIA

Characteristically in sJIA, arthritis develops weeks or even months after the onset of systemic disease with the characteristic spiking fever . Given the broad differential diagnosis of sJIA, these patients constitute a real diagnostic challenge in practice in the early phase (before the onset of arthritis), because arthritis is an obligatory feature for the clinical diagnosis of sJIA. Cytokine profiles, specifically levels of IL-18, as well as levels of S100A8-A9 and S100A12, could facilitate the diagnostic process in these patients, presenting with fever of unknown origin (Vastert et al., EULAR 2011, abstract FRI-0195).

Thus, it could be very helpful to explore the inclusion of biomarkers in future diagnostic criteria for sJIA. IL-18 is, similar to specific S100A proteins, such a promising biomarker for sJIA, not only for diagnostic purposes but also in monitoring response to treatment.

The conundrum of elevated IL-18 and depressed NK cell function in sJIA

IL-18 is a strong activating molecule for NK cell function in human physiology . However, notwithstanding the increased levels of IL-18 in active sJIA, most sJIA patients display deficits in NK cell lytic function as well as in numbers . Remarkably, NK cells from systemic-onset JIA patients fail to up-regulate cell-mediated killing molecules, such as perforin and IFN-γ after IL-18 stimulation in vivo and in vitro. This results in defective lytic function of NK cells in sJIA. Mechanistically, the apparent paradox of elevated IL-18 levels and low NK cell functionality is the consequence of a reversible phosphorylation defect of the IL-18 receptor (IL-18R) in NK cells . IL-18 stimulation in vitro of NK cells from active sJIA patients does not induce the phosphorylation of receptor-activated mitogen-activated protein (MAP) kinases. Immunoprecipitation of IL-18Rβ showed that NK cells from systemic-onset JIA cannot phosphorylate this receptor after IL-18 stimulation. This NK cell deficiency is at least partially reversible, as treatment with rIL-1RA resulted in an increase in NK cell numbers, lytic function and IFN-γ production. Altogether, the increased levels of IL-18 and depressed functionality and numbers of NK cells in sJIA are closely linked. It will be crucial to see whether, for example, rIL-1RA treatment early in the disease course may lead to a restoration of these abnormalities. If this were the case, it would offer a unique option to reset the aberrant immune response and thus to set back the clock.

Macrophage activation syndrome in sJIA: intrinsic NK cell defects?

Low NK cell function is suggested to underlie the association between sJIA and macrophage activation syndrome, MAS . MAS is considered to be an acquired form of hemophagocytic lymphohistiocytosis (HLH) and is an intriguing and potentially lethal complication of active sJIA. MAS occurs in approximately 10% of sJIA patients , whilst indications of hemophagocytosis can be observed in the bone marrow of as many as 50% of sJIA patients at the time of diagnosis . There is no other (rheumatic) disease with such a strong association with MAS . Moreover, in this respect, sJIA differs from the classical monogenic auto-inflammatory diseases or periodic fever syndromes in which NK cell function is not affected and MAS is only sporadically reported .

Interestingly, heterozygous mutations in PRF1, one of the genes involved in familial HLH (type 2 fHLH), can be linked to the occurrence of MAS in sJIA. Another study recently found an association of MAS in sJIA with polymorphisms in the gene UNC13D, the gene responsible for type 3 fHLH (when bi-allelic mutations are present) . These data point to an intrinsic defect or susceptibility of NK cells in sJIA and confirm the close relationship and clinical resemblance of the HLH syndromes with MAS, supporting the classification of MAS in sJIA as a secondary type of HLH .

Macrophage activation syndrome in sJIA: related to elevated IL-18?

Therefore, the occurrence of MAS in sJIA may in part be explained by an intrinsic NK cell susceptibility . This is not the complete picture: in addition, the immunological environment especially high IL-18 levels are influential. Indeed, elevated levels of IL-18 are present in the plasma and synovial fluid of sJIA patients, but not in polyarticular JIA. MAS is hardly seen in polyarticular JIA, whereas it is a rather frequent complication of sJIA. Moreover, hemophagocytosis in familial and acquired HLH may be both IFN-γ and IL-18 driven with a possible role for the IL-18 binding protein (IL-18BP), the natural regulatory or inhibitory protein of IL-18 . Lastly, an active sJIA patient, sampled while developing MAS, showed an even further increase in plasma IL-18 levels and a further decrease in NK cell lytic function during MAS. Altogether, these findings suggest that the strong association of MAS with sJIA is linked to the high IL-18 levels as well as depressed NK cell lytic function in sJIA.

Is IL-18 driving chronic inflammation in sJIA?

Thus, the picture that arises of sJIA from our and other studies is that of a complex auto-inflammatory IL-1-mediated disease, with IL-6- and IL-18-mediated inflammatory features as well. One of these striking features, which is not seen in other auto-inflammatory diseases, is the defective IL-18 NK cell axis. Most sJIA patients achieve inactive disease or disease remission when intervened early in the disease with rIL-1RA therapy. However, as described in other studies as well , not all sJIA patients respond equally well to rIL-1RA therapy. Apparently, the targeted blockade of IL-1R and IL-18R by rIL-1RA is insufficient in a subset of patients to prevent or overcome a perpetuating loop of chronic inflammation. Interestingly, many of these patients displayed persistently or increasing levels of IL-18 (unpublished observations). This raises the question whether persistently elevated IL-18 is one of the driving factors for chronic inflammation in sJIA and should be the focus of future studies in sJIA. Indeed, IL-18 has shown to induce Th1- as well as Th17-mediated T cell responses, probably in concert with IL-23, in published literature , and there are indications for a functional role of IL-18 in chronic joint inflammation in RA and murine models of arthritis as well .

Altogether, persistently elevated IL-18 might be involved in the perpetuation of inflammation in sJIA, shifting from primarily innate responses in the early phases of the disease to the more chronic and complex immune responses, including adaptive T helper cell-mediated responses, in (a subset of) chronically affected sJIA patients.

Is chronic sJIA an expression of defective immune regulation?

Indeed, increasingly indications are found that (chronic) sJIA can also be seen as a defect of immune regulation . Peripheral blood mononuclear cells (PBMC) of untreated sJIA patients display increased transcripts of negative regulators of innate immune responses, such as IL1RN (encoding IL-1a) and SOCS3 (encoding Suppressor of Cytokine Signalling 3) . SOCS3 is induced by IL-1 and IL-6 and negatively affects their signalling pathways. Another pathway that might be involved in the more chronic course of sJIA is the balance between regulatory T (Treg) cells and Th17 cells. Most patients who underwent autologous stem cell transplantation (ASCT) for severe JIA are sJIA patients. These severely affected, multidrug-treated patients display significantly decreased levels of Treg before ASCT, and more importantly, their response to this treatment seems to be associated with the restoration of the Treg network in peripheral blood after ASCT. Moreover, in autoimmune diseases like the other types of JIA, it has recently been shown that the balance or ratio of Treg and Th17 cells associates with the course of the disease, more regulatory cells opposed to Th17 cells being a favourable determinant of a more benign course . Interestingly, environmental factors, especially pro-inflammatory cytokines like IL-6 and IL-1β, can induce the induction of T effector (Teff) cells (especially TH17 cells) from Treg cells .

Altogether, these observations point to a complex interplay, especially in the subset of sJIA patients with a more chronic course, between innate cytokines like IL-1, IL-6 and probably also IL-18 and the balance of different T cell subsets (both Treg and Th1/Th17) affecting the course of chronic inflammatory arthritis. This should be an important focus for further study.

Translating insight into therapy for sJIA

It requires a continuous and complex effort to translate insight in immune-mediated diseases into novel therapeutic regimens. The example of sJIA shows the strength of a bench–bedside approach that translates basic research to patients. A better understanding of the IL-18 NK cell axis has directly led to exploration of rIL-1RA as first-line therapy in new sJIA patients, even before the use of systemic steroids . This strategy seems to be clinically beneficial, albeit in a small cohort of patients. Therefore, no definite conclusions can be drawn yet. The experience however shows the proof of principle on how to translate understanding into treatment. A similar approach may be possible as soon as specific antibodies for other important pro-inflammatory proteins such as S100A12 and S100A8/9 protein (complexes) become available. However, apart from treatment, the use of biomarkers to guide therapy in sJIA is already close to reality. Based on a wealth of previous studies, it seems reasonable to expect that it will be possible to taper, stop and, if needed, reinstate therapy in sJIA patients with clinically inactive disease guided by the levels of IL-18 and S100 proteins . At the least, the first preliminary observations show that this strategy seems remarkably effective in the majority of patients in the early phase of the disease.

A subgroup of sJIA patients will not respond well to IL-1 blockade and will need further optimization of therapy. This emphasizes the importance of finding ways to better select patients very early, preferably at the time of diagnosis. The crucial challenge will then be to identify who is likely to benefit from rIL-1RA therapy, and who will be better off with other biologicals like IL-6 blockade.

Ideally, this choice should be made based upon several characteristics with different dimensions. One could envision a combination of clinical features, genetic risk factors and a set of immune biomarkers that would help to define an individual risk and response to therapy profile in sJIA. Based on this profile, a choice could be made for IL-1 blockade, IL-6 blockade, or another treatment modality. IL-6 blockade could then be reserved to patients with a high risk to fail to respond to rIL-1RA monotherapy. Given the impressive therapy responses to IL-6 blockade in steroid-dependent or steroid-resistant sJIA patients, and given the fact that IL-6 blockade is already registered and licenced by the European Medicines Agency (EMA) for use in sJIA, and rIL-1RA is only permitted off label in sJIA, a randomized controlled trial including initial therapy arms with IL-1 blockade, IL-6 blockade and systemic steroids in a crossover design combined with expected genetic and immune profiling ideally would help to solve this issue. This should be executed in international collaboration, for example, by the PRINTO trial organization, probably with an innovative trial design given the rarity of this disease.

Future perspectives in sJIA research

Notwithstanding these spectacular developments in both insight and treatment options for sJIA in the past 10 years, there is still much unknown . Regarding the mechanisms of disease, the possible interplay of inflammasome-derived innate cytokines (both IL-1 and IL-18) and the development of T helper cell-mediated immune responses should be explored. Not only will dissecting this interplay result in possible new targets for therapy in sJIA but also this knowledge may extend beyond the scope of sJIA and influence other chronic inflammatory diseases of childhood as well.

Increased insight in treatment for polyarticular JIA

For the second part of this review, we focus on how current therapeutic interventions in patients with severe JIA affect the key determinants of inflammation in polyarticular JIA. It has been shown since >30 years that T cells are involved in the immune pathogenesis of JIA . In the past 15 years, the understanding on how different subsets of T cells interact in JIA has increased substantially . At least both Th1 and Th17 T cell subsets as well as different types of Treg contribute to the pathophysiology of JIA and determine its course . Dissecting the specific contributions of each subset, as well as the complex interplay between them, is important in developing new targets for therapy. However, it is crucial as well to understand how the current therapeutic modalities affect these newly identified key players in chronic inflammation in JIA. It is conceivable that with this improved insight a more targeted or fine-tuned treatment is feasible.

Treg and Teff cell subsets determine the course of JIA

The mechanisms underlying the pathogenesis and pathophysiology of polyarticular JIA are not exactly known. However, there are strong indications that the chronicity of inflammation in polyarticular JIA is the result of insufficient regulation by Treg and other regulatory mechanisms over Teff cells . We know since the early work of Sakaguchi that Treg are crucial for the regulation of inflammation and the maintenance of immune tolerance in both animals and human . In chronic inflammation in JIA, the course of oligoarticular JIA in either a remitting, benign course of the disease or a transition into a chronic course is associated with the presence of two types of Treg cells: CD4+CD25+Treg cells, expressing FOXP3mRNA in the joint , and heat shock protein (HSP)60 reactive T cells. Both cell types display a tolerogenic or regulatory profile of cytokine secretion upon stimulation in peripheral blood and synovial fluid .

At the site of inflammation, the joint itself, increased numbers of Treg cells are present when compared to peripheral blood, although these cells apparently are incapable of regulating the inflammatory loop in the joint in children with a prolonged course of their JIA. The ratio between (CD4+FOXP3+) Treg cells and Teff cells (in this study specifically Th17 cells) is correlated with the disease course in oligoarticular JIA as well . This opens the way for new therapies that might specifically target the expression of FOXP3 and/or target effector cells.

Therefore, Treg and Teff subsets and their interaction seem to be related to the disease course, at least in oligoarticular and polyarticular JIA.

How therapies increase insight in mechanisms of disease in polyarticular JIA

Thus, it seems that the balance between both auto-aggressive Teff cells and Treg are crucial in non-sJIA. On the one hand, Teff subsets, especially at the site of inflammation, could be resistant to suppression, while on the other hand Treg could have an impaired function of Treg at the site of inflammation. In appreciation of this model, therapeutic strategies should therefore not only be directed at augmenting numbers and/or function of Treg in severe polyarticular JIA patients but simultaneously also intervene in the function of Teff or in the circumstances within the joint that favour skewing of T cell responses towards Teff cells . Besides the search for novel targets of therapy, the increased insight in pathophysiology of JIA raises the question how the current therapeuticals specifically affect the Teff and Treg subsets. As discussed below, we showed different modes of action on the Treg–Teff balance of the current arsenal of therapeutic modalities for severe JIA.

MTX treatment in JIA is associated with increased, and not decreased T cell responses, but does not result in increased Treg numbers or function

For patients with polyarticular JIA and a chronic disease course, most evidence-based guidelines for therapy propose to start with a DMARD in the first weeks after diagnosis . The most widely used DMARD in JIA is MTX, generally dosed around 10–20 mg/m ² . Although MTX is already used since the late 1980s in JIA and RA, there is still debate on its exact mechanisms of action. Importantly, it takes several weeks (generally 6–12 weeks) before the beneficial effects of low-dose MTX, as used in chronic arthritis, are clinically being noticed. Although initially MTX was thought to exert its effect mainly by inducing apoptosis in Teff cells , it is currently believed that the anti-inflammatory actions of MTX are mediated via the release of anti-inflammatory adenosine. This adenosine is proposed to be released after cleavage of adenosine triphosphate (ATP) by the 5′ ectonucleotidases CD39 and CD73. CD39 is a marker for memory-type Treg, and decreased levels of C39 + Treg are associated with autoimmunity, for example, multiple sclerosis . However, the exact role of CD39+ Treg is still controversial as, for example, in JIA the numbers of CD39+ Treg are increased at the site of inflammation (synovial fluid) when compared to peripheral blood . Thus, it could be hypothesized that MTX may somehow alter the number and/or function of CD39+ Tregs. However, in polyarticular JIA, treatment with MTX does not result in an increased number of CD4+FOXP3+ Treg in peripheral blood. In addition, the expression of CD39 on Treg is unaffected by MTX treatment. Moreover, in vitro assays exploring the suppressive capacity of CD4+CD25+CD127 ^low Treg do not show improved suppression of proliferation of CD4+CD25− T cells by these Treg in patients responding to MTX treatment. Interestingly, response to MTX is also not associated with decreased Teff cell responses. Instead, it even leads to increased proliferation and maintained, not decreased, cytokine production of both in vitro-stimulated CD4+ as CD8+ Teff cells from peripheral blood. These results point to a modulatory, and not suppressive, mode of action of MTX in JIA, which does not seem to be directly mediated via Treg function. This is an important finding that clearly needs to be further explored.

TNF blockade (etanercept) targets resistance of Teff cells to Treg-mediated suppression in JIA

Currently, the use of TNF blockade in JIA is restricted to the most severely affected children with polyarticular JIA, after failure to respond to MTX . TNF blockade in chronic arthritis was developed in the final decades of the previous century, after experiments in human RA proved that TNF-α played a critical role in the production of a variety of pro-inflammatory cytokines in synovial cells . It has proven to be very effective in JIA as well, even in patients resistant to MTX and other DMARDS At this moment, two different TNF blockers are registered for use in JIA, etanercept (soluble TNF-receptor) and adalimumab (a monoclonal antibody directed to TNF-α). Most mechanistic data on the mode of action of TNF blockers are available from studies involving infliximab, the first identified monoclonal antibody to TNF-α and primarily used in adult RA. These studies suggest a critical role of TNF blockade by infliximab via modulation of numbers, subsets and function of Treg in RA .

However, the immune pathophysiology of JIA seems to differ from RA in several important aspects, for example, with respect to the role of regulatory and Teff cells . Regarding treatment, the most commonly used TNF blocker in JIA still is etanercept, not infliximab, and these agents differ, for example, in their ability to target membrane-bound TNF-α . Interestingly, etanercept does not affect the number or function of Treg in JIA directly. Instead, etanercept targets the increased phosphorylation of protein kinase B (p-PKB) in effector T cells and enhanced both in vitro and in vivo their responsiveness to suppression by Treg. Therefore, etanercept targets Teff cells at the site of inflammation in JIA, rendering Teff cytokine production susceptible for suppression again.

ASCT is resetting the immune balance and induces tolerogenic changes in autoreactive T cells

Since 1997, in an European collaboration (the European Bone Marrow Transplantation Working Party), ASCTs were performed for severe therapy-resistant JIA . These patients were resistant to all therapeutic modalities at that time, with (a high risk of) considerable disease-associated damage and suffering from (severe) side effects of chronic therapy (for example, growth retardation due to long-term use of steroids). ASCT treatment consists of intensive immune suppression through high-dose chemotherapy and/or total body irradiation (conditioning), followed by infusion of autologous bone marrow cells (transplantation of autologous stem cells), resulting in immune reconstitution in the months hereafter. The first concept of the mechanism of action of ASCT was that primarily the conditioning regime, with severe immune suppressive agents, resulted in depletion of auto-aggressive effector and memory T cells and subsequenty in clearance of chronic inflammation . However, with the recognition of the crucial role of Treg in JIA, we hypothesized that its effects were mediated via Treg as well. Indeed, it turned out that resetting the immune system by immune ablation via irradiation and chemotherapy restores the T cell balance in severe JIA. We showed preferential reconstitution of the CD4+CD25 ^bright Treg in the first 12 months after ASCT, resulting in immune reconstitution of T cells via the thymus in a tolerogenic environment. Moreover, we observed the conversion of a HSP60-peptide T cell response with an inflammatory character (low ratio of cytokines IL-10:IFN-γ, low ratio of transcription factors GATA3:Tbet) pre-ASCT, to a more tolerogenic peptide-specific T cell response post ASCT (high ratio of IL-10:IFN-γ, high ratio of GATA3:Tbet). These data indicate that ASCT affects both autoreactive (effector) T cells as well as Treg cells, when restoring the balance in severe, therapy-resistant JIA. However, the fact that a significant number of JIA patients relapsed after ASCT exemplifies that more research is needed to optimize this treatment in the near future. Several hypotheses have been explored making use of an experimental model of arthritis, showing, for example, that indeed T cells are crucial for the reconstitution of the immune balance after SCT .

The chronicity of inflammation in JIA seems determined by factors in the joint

Although sJIA has clear characteristics of a complex auto-inflammatory disease, it also has clear features in common with the other types of JIA, the most obvious one being the development of chronic arthritis. During the development of chronic arthritis, in both sJIA patients and non-systemic, polyarticular JIA patients, Teff cell responses seem to share a common phenotype. Several recent studies indicate that skewing of naïve T cells to T eff subsets (either Th1, Th17 or Th1/Th17 intermediate cells) takes place at the site of inflammation, the joint itself . Interestingly, it is getting clear now that these different T cell subsets show plasticity; in other words, T cell subsets can still be triggered to develop into other subsets. Remarkably, this induction is not only from one Teff cell subset into another Teff subset (for example Th17 to Th1/Th17 intermediate or Th1 Teff) but even from Treg into Th17 Teff cells . Most importantly, pro-inflammatory cytokines, especially IL-1β, IL-6 and possibly TNF-α as well, play a central role in the skewing of these naïve T cells into Teff cell subsets . One can hypothesize that in severe, chronic JIA circumstances within the joints are such that Teff cells are continuously being skewed and activated into resistant Teff cells .

Putting the pieces together: translate immune mechanisms into therapy for severe JIA

Clinically, the current treatment regimens in JIA can be characterized as ‘step-up’ approaches. For example, the current treatment guidelines for JIA of the American College of Rheumatology (ACR) , as well as, for example, the German consensus on treatment of JIA , start in all JIA patients of the same subset with the same class of first-line drugs. In nonresponding patients, subsequently second- or third-line drugs are being added to the regimen. Only for well-known risk factors, or in very severe onset of disease, a direct short cut to the second- or third-line class of drugs can be made .

This approach has several advantages, for example, more intensive and expensive treatment will be selected only for those patients who will not benefit (enough) of less intensive or expensive treatment modalities. However, this approach neglects the possible advantage of a so-called window of opportunity early in the disease, in which the right therapy or combination of therapy might really change the biology of the disease instead of suppressing its manifestations. This could prevent the above-described switching of T cells into certain Teff subsets that are not any longer sensitive to immune suppression by Treg. In later stages, the long-term inflammation leads to activation of multiple pathways and the release of tissue factors that altogether will render the inflammatory process refractory.

When recognizing such a window of opportunity, an alternative approach would then be the so-called ‘hit early and step down’ approach. Ideally, this new strategy would be directed by robust biomarkers, not only for prognostic purposes but also to assess subclinical disease activity to guide the start and stop strategies for different therapeutic agents. In this approach, visualized in Fig. 1 , patients with a (high risk of) a severe disease course are identified early in the disease course. In those patients, intensive and/or combined therapy can be started to induce clinical disease remission. Subsequently and based upon biomarkers that indicate sublinical disease activity, therapy will be tapered and preferably also be stopped. Alternatively, the created window of opportunity will be used to start less toxic maintenance therapy, for example, antigen-specific immunotherapy .

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Genetics, genomics, and their relevance to pathology and therapy Granulomatous inflammation: The overlap of immune deficiency and inflammation Macrophage activation syndrome and cytokine-directed therapies Assessment and treatment of pain in children and adolescents Therapeutic advancements in juvenile idiopathic arthritis A practical approach to the diagnosis of autoinflammatory diseases in childhood

Stay updated, free articles. Join our Telegram channel

Join