Adult-onset Still’s disease (AOSD) is a rare systemic inflammatory disorder that is characterised by high spiking fever, arthritis or arthralgia, and evanescent rash. Many other systemic manifestations may occur. Pathogenesis of AOSD remains partially unknown but a major role has been recently attributed to pro-inflammatory Th1 cytokines, including tumour necrosis factor-α (TNF-α), interleukin (IL)-1, IL-6 and IL-18. Despite limited evidence, mainly based on observational studies and the extrapolation to AOSD of the results of a few controlled studies that have been conducted in children with systemic juvenile idiopathic arthritis, biological agents represent a major therapeutic advances for patients with AOSD refractory to conventional treatment or presenting life-threatening manifestations. Both IL-1 and IL-6 blockade may be more effective than TNF-α blockers. Although debatable, therapeutic strategies are proposed.
- •
Pathogenesis of AOSD remains unknown but a pivotal role has been attributed to pro-inflammatory cytokines, including tumour necrosis factor-α (TNF-α), interleukin (IL)-1, IL-6 and IL-18, and innate immunity deficiency.
- •
The majority of patients with AOSD will require corticosteroids. In the presence of corticosteroid resistance or dependence, a disease-modifying anti-rheumatic drug should be used and methotrexate is the treatment of choice.
- •
In patients with refractory or life-threatening AOSD, biological agents should be considered off-label.
- •
IL-1 (anakinra) and IL-6 (tocilizumab) blockers seem to be more effective than anti-TNF-α (infliximab, etanercept or adalimumab). TNF-α blockers and anti-IL-6 may be more useful in chronic polyarthritis. Anti-IL-1 and anti-IL-6 may be more effective in the presence of systemic manifestations.
- •
Progress in the understanding of the pathogenesis of adult-onset Still’s disease (AOSD) is necessary and will help to develop new therapeutic strategies.
- •
Controlled clinical trials should be encouraged despite the rarity of AOSD.
- •
Development of new inhibitors of pro-inflammatory cytokines, including IL-18 may add new therapeutic options.
- •
Refinement of therapeutic strategies according to the presentation of AOSD is necessary.
Introduction
Adult-onset Still’s disease (AOSD) is a rare multi-systemic inflammatory disorder that was described for the first time in 1971 by E.G. Bywaters who reported a case series of 14 young adult females who presented with arthritis and systemic manifestations that were identical to those of the systemic form of juvenile idiopathic arthritis (JIA) that was identified by G.F. Still in 1897 .
Typically, the clinical features associate a high spiking fever, arthralgia or arthritis and an evanescent salmon-pink maculo-papular skin rash that is predominantly found on the trunk and proximal limbs. Many other systemic features may occur and some of them may be life threatening. The main laboratory features include a striking leucocytosis with a high percentage of polymorphonuclear and an increase of acute-phase reactants, despite a negative infectious work-up. Except for the common presence of a high serum level of gammaglobulins, there is no evidence of autoimmunity and AOSD is not associated with the presence of immunoglobulin M (IgM) rheumatoid factor, anti-cyclic citrullinated peptide or antinuclear antibodies. Despite the diagnostic value attributed to an elevated serum ferritin level – frequently discordant with that of acute-phase reactants – associated with a decrease in its glycosylated fraction (<20%), the diagnosis of AOSD remains one of exclusion. The classification criteria proposed by Yamaguchi et al. published in 1992 are widely used .
Outcome of AOSD is unpredictable but approximately 70% of the patients will develop an intermittent form of the disease or a chronic course that manifests mainly with a chronic polyarthritis and eventually joint damage .
The treatment of AOSD remains largely empiric based on the results of single case reports or small case series. The rarity of AOSD makes it difficult to carry out controlled clinical trials and none have been conducted in adults . In the years following its description, most of the patients with AOSD have been treated as children having the systemic form of the JIA with high-dose aspirin or other non-steroidal anti-inflammatory drugs (NSAIDs). However, only 15–20% of the patients with AOSD respond to this treatment and in the majority corticosteroids are necessary to control both the systemic and articular manifestations of the disease . In the intermittent and chronic forms of the disease most of the disease-modifying anti-rheumatic drugs (DMARDs) that have been proposed in rheumatoid arthritis were used in AOSD. Despite the absence of controlled clinical trials, methotrexate (MTX) is clearly the drug that emerged as the most useful DMARD in association with corticosteroids. It has been shown to be effective in approximately 70% of the patients and it has a corticosteroid-sparing effect .
More recently biological agents showed promising results in the treatment of some patients with AOSD. This chapter reviews the literature regarding the use of biological treatment in AOSD.
Rationale to use biological agents in AOSD
The pathogenesis of AOSD is still unknown and remains the subject of much research . However, several studies have suggested that an abnormal production by uncontrolled, activated T cells and macrophages of Th1 pro-inflammatory cytokines, including interleukin (IL)-1, IL-6, IL-18, tumour necrosis factor-alpha (TNF-α) and interferon-gamma, plays a pivotal role in AOSD .
The common intermittent disease course, the high serum levels of IL-1 and IL-18, and the shared clinical and laboratory manifestations of AOSD with several genetic auto-inflammatory syndromes (especially the cryopyrinopathies and the familial Mediterranean fever) have also contributed to consider the role of innate immunity and the inflammasome in AOSD. Yet, AOSD has been classified as a non-Mendelian, multifactorial auto-inflammatory syndrome .
These data were the basis of the off-label use of the biological agents that were originally developed for rheumatoid arthritis (RA), in patients with AOSD refractory to conventional therapy.
Rationale to use biological agents in AOSD
The pathogenesis of AOSD is still unknown and remains the subject of much research . However, several studies have suggested that an abnormal production by uncontrolled, activated T cells and macrophages of Th1 pro-inflammatory cytokines, including interleukin (IL)-1, IL-6, IL-18, tumour necrosis factor-alpha (TNF-α) and interferon-gamma, plays a pivotal role in AOSD .
The common intermittent disease course, the high serum levels of IL-1 and IL-18, and the shared clinical and laboratory manifestations of AOSD with several genetic auto-inflammatory syndromes (especially the cryopyrinopathies and the familial Mediterranean fever) have also contributed to consider the role of innate immunity and the inflammasome in AOSD. Yet, AOSD has been classified as a non-Mendelian, multifactorial auto-inflammatory syndrome .
These data were the basis of the off-label use of the biological agents that were originally developed for rheumatoid arthritis (RA), in patients with AOSD refractory to conventional therapy.
Biological agents used in AOSD
TNF-α blockers
The usefulness of TNF-α blockers was originally suggested by Stambe and Wicks who reported in 1998 the efficacy of thalidomide, a weak inhibitor of TNF-α, in a patient with AOSD resistant to conventional treatment, despite the absence of increased serum TNF-α level in this patient .
Infliximab
Infliximab has been the first TNF-α blocker to be used in AOSD. It is a murine–human chimaeric monoclonal antibody that binds to soluble TNF-α and can also inhibit the interaction of TNF-α with its cellular receptors. As in RA, infliximab has been prescribed at a dose of 3–5 mg kg −1 body weight given intravenously at weeks 0, 2 and 6, and every 6–8 weeks thereafter.
In a small case series of six AOSD patients reported by Kraetsch et al. , infliximab was credited in all patients of a marked efficacy, on both systemic and articular manifestations. The effect was usually rapid, sometimes after the first intravenous infusion, and sustained with a maximum follow-up of 28 months in this series. Infliximab was associated with MTX and azathioprine in one patient each, and infliximab has a corticosteroids-sparing effect. In two other small case series adding up seven patients with active AOSD resistant to MTX, infliximab combined with MTX in all but one patient was rapidly effective . Again, remission was sometimes obtained after a single infusion and interested both systemic and joint manifestations. In a few patients, systemic or joint manifestations resumed while on maintenance therapy with infliximab and could eventually be controlled with a more frequent infusion regimen (every 4 weeks) . A few other single case reports have also highlighted the usefulness of infliximab in AOSD .
However, these spectacular results obtained with infliximab should be weighted with the results of the retrospective study published by the French ‘Club Rhumatismes et Inflammation’ group on all prescriptions of anti-TNF-α (infliximab and etanercept) treatment in AOSD patients who were resistant to conventional DMARD therapy . The majority of the 20 reported patients had a chronic disease course and the mean disease duration was 8.5 years (range 2–21 years). All of them had previously received MTX and several other DMARDs including cyclophosphamide in four and azathioprine in three. In these highly selected AOSD patients, disease expression was mainly articular (15 out of the 20 patients had a polyarticular chronic disease course). Remission was defined as a complete resolution of all clinical and laboratory features of AOSD; except for joint damage, failure was defined as an absence of significant improvement within 1–3 months after initiation of TNF-α blocker, and partial remission was defined as the persistence of one or several AOSD related manifestations. Among the 15 patients who were treated with infliximab (some patients successively received both infliximab and etanercept), 4 out of 15 (27%) and 9 out of 15 (60%) achieved a complete and a partial response, respectively. In the two remaining patients infliximab therapy was ineffective . At the time of the survey, 10 out of the 15 patients treated with infliximab had discontinued the treatment after a mean treatment duration of 9 months because of a lack of efficacy (six patients) or a side effect (three patients).
Overall, infliximab is well tolerated and the side effect profile is identical to that reported in RA and includes mainly infusion reaction and skin rash . However, one AOSD patient who had a positive hepatitis B virus surface antigen developed a fulminant hepatitis 2 weeks after receiving a second infliximab infusion and required a liver transplantation . Another patient with inactive carriage of hepatitis B virus antigen and low hepatitis B-DNA level at polymerase chain reaction (PCR) assay had no exacerbation of hepatitis during infliximab therapy .
Etanercept
Etanercept is a recombinant form of the human 75-kd TNF receptor fusion protein. It has been used in a prospective 6-month duration open-label trial in 12 AOSD patients (10 women and two men) who had active chronic polyarthritis and a mean disease duration of 10.8 years . All these patients were resistant to conventional therapy and 10 of them had received or were receiving MTX. Etanercept was prescribed at the classic dose of 25 mg twice weekly subcutaneously. Response to treatment was assessed according to the American College of Rheumatology (ACR) improvement criteria . At 6 months, seven out of the 12 patients achieved an ACR-20 and among them, four and two achieved an ACR-50 and an ACR-70, respectively. Only three patients had concomitant systemic manifestations and only one of them improved; arthritis did not improve in any of these three patients. For four patients the etanercept dose had to be increased to 25 mg 3 times per week in the absence of clinical response by 8 weeks but then only one patient responded. Two patients discontinued etanercept because of disease flare.
A few other single case reports of AOSD patients treated with etanercept have been published, and as the result of a classic publication bias most of these reported a favourable effect of the TNF-α blockade . In the patient reported by Asherson and Pascoe, etanercept was the only drug that was able to control a severe case of AOSD complicated with cardiac amyloidosis . In another patient, etanercept was also able to stop the progression of renal amyloidosis . Despite the risk for etanercept like infliximab to exacerbate cardiac failure , etanercept was used successfully in two patients with AOSD complicated by myocarditis and heart failure .
In the retrospective study conducted by the French ‘Club Rhumatismes et Inflammation’ group, among the 10 patients who were treated with etanercept, only one achieved a complete response and seven out of 10 a partial response . Further, in the more recent studies addressing the effect of the newer biological treatment (see below) most of the reported patients had failed with etanercept or infliximab therapy.
Systemic JIA is the paediatric counterpart of AOSD and it is reasonable to extrapolate the results obtained in children where a few controlled clinical trials are available . Interestingly, in the withdrawal trial of etanercept in JIA, 17 children with systemic JIA whose disease responded in the first part of the trial were randomised to continue etanercept (nine children) or to receive placebo (eight children); 44% experienced a disease flare in the etanercept group compared to 88% in the placebo group .
Adalimumab
Adalimumab has been effective in some patients with AOSD . In the patient reported by Agarwal et al., the disease course on adalimumab was complicated by the occurrence of disseminated histoplasmosis with macrophage activation syndrome that mimicked an AOSD systemic flare .
Altogether more than 120 patients treated with a TNF-α blocker have been published including those in whom it was ineffective . Switching from a TNF-α blocker to another one may be useful , and although head-to-head comparison is not available, infliximab may be more effective than etanercept .
Anakinra and the other inhibitors of IL-1
Anakinra (IL-1Ra) is a recombinant inhibitor of the IL-1 receptor that blocks the action of IL-1 that was originally developed and approved for the treatment of RA. Evidence of the role of the pro-inflammatory IL-1 cytokine in the pathogenesis of AOSD and the dramatic effect of anakinra in familial cold autoinflammatory syndromes was appealing to its use in AOSD .
Fitzgeral et al. in 2005 were the first to use anakinra in AOSD . In this open-label study on four patients with AOSD refractory to MTX and also to etanercept in two, treatment with anakinra resulted in a rapid and complete resolution of both systemic and articular manifestations that was sustained during a 6–14-month follow-up .
Several single case reports have emphasised the spectacular effect of anakinra in AOSD , sometimes in patients having life-threatening manifestations . Two additional small case series of four patients each reported the efficacy of anakinra in AOSD patients resistant to conventional treatment, including TNF-α blockers in two. In the study by Kötter et al. , during a 10-year period 14 patients were followed-up in a single rheumatology department and four patients who were resistant to conventional treatment were treated by anakinra. Two out of these four patients had elevated IL-1β serum levels that normalised with anakinra; serum IL-6 was elevated in only one patient and TNF-α level was slightly elevated in all four patients. IL-18 serum level was very high and returned to normal on anakinra in all four patients.
The French experience of the treatment of patients with AOSD and children with systemic onset JIA with anakinra has been reported by Lequerré et al. . A total of 35 patients were included in this retrospective survey, including 15 with AOSD. Their mean age was 38.1 years, the mean disease duration was 7.8 years, all 15 had active arthritis and 13 had also systemic manifestations. All 15 patients had received or were receiving MTX, and 10 (67%) had also been treated with a TNF-α inhibitor that was not effective. Eleven patients out of the 15 had a rapid and major improvement and were still receiving anakinra at the latest follow-up (11–27 months). Among those 11 patients, nine achieved a complete response (no systemic symptoms and at least an ACR-50 score improvement) and two additional patients obtained a partial response (no systemic symptoms and 20–49% improvement in the ACR score). The dose of corticosteroids could be largely diminished (mean daily dose of 26.8–8.6 mg). Four patients discontinued the anakinra because of lack of efficacy or side effect.
Overall, despite the lack of randomised controlled trials in AOSD, anakinra seems to have an impressive efficacy that is rapid and sustained, and associated with a corticosteroid-sparing effect in most patients . Inhibition of IL-1 with anakinra might be more effective in patients with highly active systemic disease than in patients with isolated chronic arthritis .
Recent results obtained in children with systemic JIA treated with anakinra also give support to the usefulness of a biologic agent in AOSD, despite its less impressive efficacy in children. Indeed, in a recent multicentre, randomised, double-blind placebo controlled trial in patients with systemic JIA (ANAJIS trial), efficacy of anakinra was demonstrated at least in the short term : at 1-month, eight children out of 12 (67%) in the anakinra group were responders compared to one child out of 12 (8%) in the placebo group ( p = 0.003). Established systemic JIA may be less responsive with incomplete response to anakinra as suggested by the survey by Nigrovic et al., who showed a rapid resolution of systemic symptoms and prevention of refractory arthritis in almost 90% of systemic JIA with anakinra as first-line therapy . Whether this ‘window of opportunity’ could be identical in AOSD remains to be demonstrated.
Anakinra is usually well tolerated except for a common self-limited erythema at the injection site . Other side effects are uncommon but at least one patient with a severe systemic inflammatory response syndrome and an acute adult respiratory distress syndrome has been reported . Further, it should be emphasised that a macrophage activation syndrome has been reported in a few patients while receiving anakinra . Although anakinra is firmly contraindicated during pregnancy and breastfeeding, an uneventful outcome was reported in one patient .
Anakinra is usually prescribed at the dose of 100 mg daily subcutaneously and is commonly associated with MTX . A 200-mg daily dose of anakinra has been used in one patient . Treatment with anakinra may be purely suspensive as even a brief interruption of treatment is frequently associated with a disease flare . Duration of treatment with anakinra has not been defined. However, it is reasonable to gradually increase the time interval between injections after a prolonged period of remission . The structural effect of anakinra in chronic articular form of the disease remains unknown. In one patient, despite rapid and sustained improvement in clinical and laboratory parameters on anakinra joint damage developed .
Anakinra has a very short half-life of 4–6 h that requires daily subcutaneous injection that is quite inconvenient. Recent development of canakinumab, a fully human monoclonal antibody against IL-1β with a half-life of 26 days, is promising. It has been approved for the management of cryopyrin-associated auto-inflammatory syndromes due to mutations in the NALP3 inflammasome and its long half-life makes it possible to administer it every 8 weeks . It has been used successfully in two patients with AOSD who were resistant to anakinra . Importantly, canakinumab has been shown to be very effective in children with systemic JIA. In a phase II multicentre, open-label study, 23 children with active systemic JIA were administered a single subcutaneous dose of canakinumab and were re-injected only upon disease relapse; among the 15 responders (60%), the response was sustained over time and corticosteroid dosage could be decreased .
Rilonacept (IL-1 trap molecule, a soluble dimeric fusion protein) is another long-acting inhibitor of IL-1 that is administered as a weekly subcutaneous injection (160 mg weekly after a first loading dose of 220 mg). It has been used successfully in four patients with AOSD .
Inhibitors of IL-6
IL-6 was also considered as a suitable target for the treatment of AOSD as its serum level is commonly markedly elevated in active disease .
Tocilizumab is a humanised anti-IL-6 receptor antibody that recognises both the membrane-bound and the soluble form of IL-6 receptor, and specifically blocks the actions of IL-6. Several case reports have shown promising preliminary results of tocilizumab in AOSD . In a small case series of three patients who were refractory to many medications including MTX (3/3), anti-TNF-α (2/3) and anakinra (3/3), tocilizumab was rapidly effective . A recent Japanese survey reported that 11 patients treated with tocilizumab achieved a good response often after a switch from a TNF-α blocker or anakinra because of failure . In one patient tocilizumab was discontinued and the patient remained in remission for over 8 years.
The French experience of 14 patients (nine women and five men, mean age of 38.4 years, disease duration of 13.6 years) with refractory AOSD treated with tocilizumab was also recently reported . All these patients had persistent active arthritis (14/14) or systemic manifestations (7/14) that were resistant to conventional treatment including MTX (14/14), anakinra (14/14) and anti-TNF-α drugs (12/14). Response to tocilizumab was objectively assessed using the Disease Activity Score-28 (DAS-28) and the European League against Rheumatism (EULAR) criteria, and systemic improvement was defined as the resolution of systemic manifestations. In most patients tocilizumab was administered at the dose of 8 mg kg −1 every 4 weeks intravenously (recommended regimen in RA), but four patients received 8 mg kg −1 every 2 weeks. Tocilizumab was combined with MTX in 8 patients and with leflunomide in 2. At 6 months, the DAS-28 decreased from 5.61 to 2.91 and a EULAR remission (DAS < 2.6) was achieved in 57% (8/14) of the patients, and resolution of systemic manifestations was observed in 86% (6/7). In most patients, response to tocilizumab was rapid and sustained. Tocilizumab had also a corticosteroid-sparing effect and prednisone dose could be decreased from a mean baseline daily dose of 23.3–10.3 mg at 6 months. The safety profile appeared acceptable; one patient developed a necrotising angiodermatitis and another had chest pain and chills at each tocilizumab infusion.
In addition to these observational and uncontrolled studies in AOSD, a randomised placebo-controlled trial demonstrated the efficacy of tocilizumab in children with systemic JIA . Fifty-six children with systemic JIA refractory to conventional treatment were given three doses of tocilizumab (8 mg kg −1 every 2 weeks) during an open-label lead-in phase. Then, children who responded to tocilizumab were randomised to receive placebo or to continue the active drug for 12 weeks. Among the 43 patients who responded to the lead-in phase and were randomised to the double-blind phase, 4/23 (17%) in the placebo group maintained an ACR Pedi 30 response and a CRP (C-reactive protein) <15 mg l −1 compared with 16 (80%) of the 20 patients who received the tocilizumab ( p < 0.0001).
Despite the limited number of patients (<50 patients) with AOSD treated with tocilizumab, its safety profile seems identical to that reported in RA. Of note, two patients while on tocilizumab developed a macrophage activation syndrome, one being related to cytomegalovirus infection . Further, it should be kept in mind that anti-IL-6 directly suppresses acute-phase reactant production such as CRP, even in the presence of a bacterial infection.
Intravenous immunoglobulins
Intravenous immunoglobulins (IVIgs) have been proposed in AOSD at the classical dose of 2 g kg −1 administered in 2–5 days every month. In two open-label studies, they have been credited to be effective in eight patients out of 14, early in the disease course . In a few other single case reports the efficacy of IVIg has also been reported . It may be worth to try them in life-threatening manifestations as it was reported in a fulminant myocarditis . However, no controlled study is available in AOSD and it should be mentioned that a controlled study in systemic JIA was unable to demonstrate the efficacy of this treatment . Failure to respond to IgIV in AOSD has also been reported .
Other biological agents
Abatacept is a modulator of T-lymphocyte activation that has been approved for the treatment of RA. This agent has been used in four patients with AOSD and has been effective in two .
Rituximab, a monoclonal anti-CD20 antibody has been tried in four patients but was not useful .
Related posts:
A variety of diseases with a common trunk
IgG4-related diseases
Nephrogenic systemic fibrosis: A systemic fibrosing disease resulting from gadolinium exposure
Catastrophic antiphospholipid syndrome (CAPS)
Multicentric reticulohistiocytosis and fibroblastic rheumatism
Autoinflammatory syndromes and cellular responses to stress: pathophysiology, diagnosis and new treatment perspectives
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
