The Asia‐Pacific League of Associations for Rheumatology Consensus Recommendations on the Management of Systemic Juvenile Idiopathic Arthritis (Juvenile Still's Disease)

ABSTRACT

Background

The unique characteristics of the Asia Pacific called for the development of pragmatic and viable guidelines for the management of juvenile idiopathic arthritis (JIA). These consensus recommendations represent the second part of the guideline intending to offer an updated and regionally relevant framework for the management of systemic JIA (sJIA)/juvenile Still's disease.

Methods

A multidisciplinary task force of 35 members from 14 countries, including pediatric and adult rheumatologists as well as patient representatives, was convened. The guideline development followed the GRADE, ADAPTE, and AGREE II frameworks. Relevant international guidelines were critically appraised, and a systematic literature review was performed to address 10 PICO questions. Draft statements were discussed and voted upon using a modified Delphi process, with consensus defined as ≥ 80% agreement.

Results

Four overarching principles and eighteen statements, including five statements addressing macrophage activation syndrome (MAS) were formulated. The recommendations align with international guidelines, advocating early use of IL‐1/IL‐6 inhibitors, avoidance of glucocorticoid monotherapy in sJIA without MAS, promoting tapering of glucocorticoids once inactive disease is achieved, alongside class switching for biologic refractory cases and a treat‐to‐target approach akin to EULAR/PReS guideline. Additional features include recommending conventional synthetic DMARD when biologics are unavailable and listing the options, acknowledging regional health system heterogeneity in the Asia Pacific, mentioning tapering strategy, discouraging NSAIDs as initial monotherapy, highlighting non‐biologic salvage options, and placing less emphasis on sJIA‐associated lung diseases.

Conclusion

These recommendations provide direction for practitioners caring for sJIA patients in limited resource areas to avoid treatment delay, hence improve overall outcomes. A shared decision approach and treat‐to‐targets are emphasized.

1. Introduction

Juvenile idiopathic arthritis (JIA), one of the most common childhood‐onset rheumatic diseases, was categorized into 7 subtypes according to the International League of Associations for Rheumatology (ILAR) classification criteria [1]. Systemic JIA (sJIA) represents a distinct subtype of JIA characterized by systemic inflammation, quotidian fever, evanescent rash, and potential internal organ involvement with arthritis manifesting either initially or later in the disease course [2]. Unlike other JIA subtypes, sJIA possesses significantly higher morbidity due to complications like macrophage activation syndrome (MAS), lung disease, growth impairment, and long‐term disability. While overall mortality in JIA is rare, sJIA carries a disproportionately elevated risk of death, largely attributable to MAS and severe systemic disease activity [3]. Compared to the 10%–15% found in the West, the prevalence of sJIA is notably higher in various regions in the Asia Pacific, accounting for up to 30%–50% of all JIA cases [4, 5, 6]. Similarly, the reported prevalence of MAS is higher in this region (15%–20% and in some JIA cohorts, up to 50%) than in the West (10% or less) [2, 7, 8]. Over the past decade, there has been a notable improvement in the mortality rate associated with MAS, decreasing from 20%–30% to 8%–12%, primarily attributed to early recognition and the prompt implementation of biologics [9]. However, such improvement has not been observed in our region, where the mortality rate remains elevated at 15%–20% and in some areas, up to 33% [8, 10]. In addition to restricted availability of care and medication, particularly biologic therapies, delayed diagnosis and inadequate disease management contribute to unfavorable outcomes, as the majority of cases are managed by non‐pediatric rheumatologists [11]. The international recommendations advocated for the prompt initiation of biologic therapy, such as IL‐1 inhibitors (IL‐1i) or IL‐6 inhibitors (IL‐6i), in the therapy of sJIA [12, 13]. Nevertheless, findings from our recent regional survey indicate that IL‐1i were unavailable in the majority of participating countries, with over half reporting no access to these agents [11]. Additionally, IL‐6i were not universally accessible across all surveyed nations [11]. A practical and regionally relevant treatment guideline is necessary to enhance sJIA outcomes in the Asia Pacific, accounting for unique regional characteristics such as healthcare access disparities, cultural diversity and health perceptions, medication adherence and significant variation in clinical practice, especially among the non‐pediatric rheumatology healthcare providers.

In response to region‐specific issues, the Asia‐Pacific League of Associations for Rheumatology (APLAR) formulated evidence‐based consensus recommendations aimed at revising and refining the guidelines published in the prior 5 years. These recommendations constitute the second segment of the three‐part APLAR consensus guidelines for the management of JIA, focusing on sJIA with and without MAS. This document is designed to provide guidance for healthcare practitioners in the Asia‐Pacific region who directly manage patients with JIA. The intended audience encompasses specialists, general practitioners, specialty nurses, and other allied health professionals, who will collectively be referred to as “practitioners”. In addition, the recommendations are relevant to other stakeholders including patients, their caregivers, and health policy makers. It is crucial to understand that these consensus recommendations are not intended to prescribe a definitive course of care for any individual patient. Instead, therapeutic decisions must be informed by the unique clinical circumstances of each patient and involve shared decision‐making between the treating practitioner and the patient/caregiver, ensuring that all choices are made with the patient's best interests in mind.

2. Materials and Methods

This consensus recommendation development process followed the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology (www.gradeworkinggroup.org), which involves the process of rating the quality of the best available evidence and developing health care recommendations following the approach proposed by the GRADE Working Group [14] and adhered to the Appraisal of Guidelines for Research and Evaluation II (AGREE II) criteria [15] for consensus recommendation quality and ADAPTE Collaboration methodology for existing guideline adaptation (www.g‐i‐n.net) [16].

2.1. Task Force Working Group Organization

The APLAR Pediatric Rheumatology Special Interest Group (PR SIG) established the Task Force (TF) working groups in October 2023 (Appendix S1). There were five working groups: (1) Steering Committee (SC), chair (TA), nominated by PR SIG convenor (SFR), formed TF SC, consisting of 10 expert pediatric rheumatologists and one adult rheumatologist from 10 countries (median years of experience 19.0 years [interquartile range (IQR) 16.0–25.0]). This committee identified and appraised existing guidelines, formulated and finalized clinically relevant Patient/Population, Intervention, Comparison, and Outcomes (PICO) questions, ranked outcomes, supervised and coordinated the entire development process (Figure 1) and drafted recommendation statements and manuscripts; (2) Guideline Development Methodologist (LD), advised and guided methodology/technical process, checked evidence summaries for correctness of interpretation of evidence, and provided technical assistance to other working groups; (3) Evidence Review Panel (ERP), consisting of 13 pediatric rheumatologists, searched and identified relevant literature, critically appraised and rated the certainty of evidence; (4) Consensus Panel (CP), consisting of 14 pediatric rheumatologists, three adult rheumatologists (median years of experience 16.5 years [IQR 11.3–18.8]), and two JIA patient and parent panel (PPP) representatives from 14 countries, who assisted in formulating and refining PICO questions, identified and prioritized the critical and important outcome measures prior to finalizing the questions, such that stakeholders' values and preferences could be incorporated, reviewed evidence summaries and voted on the recommendation statements for which the PPP representatives votes carried equal weight to those of other panel members.; (5) Patient and Parent Panel (PPP), consisting of nine adult JIA patients and seven parents from six countries, provided input and opinion on benefits and adverse events of drugs, reviewed evidence summary and assisted in rating outcomes and provided input on patient values and preferences related to treatment options, outcomes and evidence. All task force members declared conflict of interest prior to all consensus statement development activity and > 95% of members had no conflict of interest. Panel members with conflicts of interest shall be excluded from all related activities, including evidence appraisal, group discussions, voting sessions, and leadership roles concerning the medications in question.

FIGURE 1.

Flowchart of consensus recommendation statement development. Briefly, The Steering committee (SC) defined the scope, drafted PICO questions and appraised the existing guidelines. The consensus panel (CP) and evidence review panel (ERP) reviewed and refined the PICO questions. The ERP reviewed and graded the certainty of evidence and drafted recommendation statements for CP members to vote on using modified Delphi process with a final of 18 consensus recommendations. APLAR, Asia Pacific League of Associations for Rheumatology.

2.2. Existing Guidelines Adaptation

The ADAPTE framework was adopted by the SC with the intention to adapt international guidelines for the current consensus recommendations [16]. The Appraisal of Guidelines for REsearch & Evaluation (AGREE) II instrument (www.agreetrust.org) was used to appraise the quality of two international guidelines which were published in the past five years [12, 13]. Three SC members appraised each guideline. It was decided by SC to set the prioritized domain (domain 3: rigor of development) score of > 70% as the threshold for high‐quality guidelines to be adapted for the current project. Both guidelines were selected for adaptation with modification [12, 13].

2.3. Consensus Development

Figure 1 summarizes the consensus recommendation development process. In brief, SC drafted the overarching principles and the first version of the PICO questions (based on predefined clinical priorities, unmet regional needs and stakeholder surveys) and determined and ranked the outcomes prior to the CP and PPP review and then finalized. The scope of these consensus recommendations was decided to be limited to sJIA by EULAR/PReS recommendations' definition as follows [13]:

• Fever is typically spiking with temperature ≥ 39°C (102.2°F) for at least 7 days.

• Rash is transient and often coincides with fever spikes, preferentially involving the trunk. It is typically erythematous (salmon pink), but other rashes (e.g., urticaria) may be consistent with the diagnosis.

• Musculoskeletal involvement is usually present with arthralgia/myalgia. Overt arthritis is supportive but not necessary for diagnosis and may appear later.

• High levels of inflammation are typically identified by neutrophilic leucocytosis, increased serum C‐reactive protein (CRP), and ferritin.

It includes patients with and without MAS, both overt and subclinical disease, and excludes patients from other ILAR categories and children with uveitis that may influence treatment decisions. The MAS is defined using Ravelli criteria as follows [17]:

A febrile patient with known or suspected sJIA is classified as having MAS if the following criteria are met:

• Ferritin > 684 ng/mL and any 2 of the following:

• Platelet count ≤ 181 × 10⁹/L

• Aspartate aminotransferase > 48 Units/L

• Triglyceride > 156 mg/dL

• Fibrinogen ≤ 360 mg/dL

Moreover, persuaded by the EULAR/PReS recommendations that sJIA and adult‐onset Still's disease are the same disease supported by compelling evidence, the TF shall henceforth utilize the term ‘Still's disease’ to refer to sJIA throughout the document [13]. Additional information regarding the PICO structure is detailed in Appendix S2.

A systematic literature review was performed by the ERP members, following Preferred Reporting Items for Systematic Reviews and Meta‐Analyzes (PRISMA) guidelines and registered on PROSPERO (CRD420251055826) [18]. The literature searches for each PICO question were conducted across MEDLINE, Embase, Web of Science, Scopus, and the Cochrane Central Register of Controlled Trials (CENTRAL) from inception through 1 April 2025, with updates on 1 September 2025 (Appendices S3 and S4; Figure S1). Because the current consensus recommendations were adapted from existing international guidelines approved by the SC using the AGREE II tool, the final evidence review excluded studies that had been incorporated into previous guidelines. The risks of bias and certainty of evidence (CoE) assessment were performed following the Cochrane Risk of Bias 2 (RoB2) tool and the GRADE approach, respectively [14, 19]. The ERP drafted recommendation statements (RS) considering the CoE from new evidence in addition to the strength of recommendation (SoR) from existing guidelines. The RS were subsequently refined and finalized by the SC. There are four OPs and a total of 18 statements including 13 for Still's disease without MAS and 5 for that with MAS.

A modified Delphi process was used to develop consensus through one face‐to‐face meeting followed by one virtual meeting. Consensus for each recommendation statement was determined through anonymous dichotomous voting (“agree” or “disagree”). Consensus was predefined as agreement by at least 80% of the CP members. If consensus was not achieved, anonymized feedback and reasons for disagreement from the preceding round were reviewed, followed by discussion to refine the statement. The revised statement was then subjected to a further round of anonymous voting. After the final recommendation statements were agreed upon, members of the SC and CP independently rated their level of agreement (LoA) with each overarching principle and recommendation statement on a scale from 0 (completely disagree) to 10 (completely agree). The mean and standard deviation (SD) of the LoA scores were calculated and reported.

Three categories of SoR were defined. A strong recommendation (using the terms “recommend” or “should”) indicated that the TF was confident that the desirable effects of an intervention outweighed its undesirable effects, or vice versa. A weak recommendation (using the terms “suggest”, “may”, or “might”) indicated that the desirable effects were likely to outweigh the undesirable effects, or vice versa; hence each patient's circumstances, values, and preferences should be considered carefully. A shared decision‐making discussion between the patient/caregiver and the treating practitioner is crucial for interventions with a weak SoR. Where evidence was insufficient to support either a recommendation for or against an intervention, no SoR was assigned.

2.4. External Review

These consensus recommendations were presented in public forums at the 28th Asia‐Pacific League of Associations for Rheumatology Congress (APLAR 2026, Seoul, Republic of Korea), the Taiwan Academy of Pediatric Allergy, Asthma, Immunology & Rheumatology, and the 2nd Pediatric Rheumatology School in Taipei, Taiwan for external review; the first draft was also sent to two pediatric rheumatology experts (see Acknowledgement) for external review.

2.5. Update

Acknowledging the fast‐paced advancements in Still's disease research and management, the TF will undertake a revision of these consensus recommendations 5 years after their publication.

3. Results

There are four overarching principles (OP) and thirteen statements for Still's disease without MAS and five statements for Still's disease with MAS. Nonetheless, the majority of statements constitute weak recommendations due to the low to very low certainty of supporting data, accounting for 94%.

3.1. Overarching Principles (Table 1)

TABLE 1.

APLAR consensus statements on systemic JIA/Still's disease management.

Recommendation statements		SoR	CoE	%A	LoA
Overarching principles
1.	Prompt diagnosis and treatment initiation of Still's disease is essential to achieve better outcomes while preventing long‐term damage.	NA	NA	100.0%	10.0 (0)
2.	Treatment target and treatment plan are based on a share‐decision between the patient/caregiver and the treatment team.	NA	NA	100.0%	9.9 (0.3)
3.	A treat‐to‐target approach is essential, with regular assessment of disease activity and timely modification of therapy, aiming for drug‐free remission, when possible, or clinically inactive disease with minimal therapy.	NA	NA	100.0%	10.0 (0)
4.	Long‐term systemic glucocorticoid therapy is discouraged due to its long‐term adverse events.	NA	NA	100.0%	10.0 (0)
The following recommendation statements apply to children and young adults with treatment naïve, newly diagnosed active Still's disease without MAS
1.	We suggest against NSAIDs as initial monotherapy	2	D	93.8%	9.5 (0.7)
2.	We suggest against systemic glucocorticoids as monotherapy. However, systemic glucocorticoids may be used as a bridging therapy.	2	D	100.0%	9.2 (0.9)
3.	If possible, we suggest initiating treatment with IL‐6 inhibitors or IL‐1 inhibitors. However, conventional synthetic DMARDs may be considered if IL‐6 inhibitors or IL‐1 inhibitors are not available.	2	B	93.8%	9.6 (0.6)
4.	When conventional synthetic DMARDs are being considered as initial therapy, methotrexate may be considered for patients with predominant articular disease, while cyclosporin may be considered for those with predominant systemic disease. In some cases, a combination of the two may also be considered. This strategy requires careful toxicity monitoring.	2	C (MTX) D (CsA)	100.0%	9.6 (0.8)
5.	We recommend against using TNF inhibitors as initial therapy.	1	D	93.8%	9.7 (0.6)
The following recommendation statements apply to children and young adults with Still's disease without MAS who do not respond to initial therapy
6.	In patients initiated on an IL‐6 inhibitor or IL‐1 inhibitor at their respective therapeutic dose (or after dose optimization), switching to either IL‐1 inhibitor or IL‐6 inhibitor is recommended, if feasible.	1	C	100.0%	10.0 (0.2)
7.	In patients initiated with a conventional synthetic DMARD at a therapeutic dose (or after dose optimization), if possible, switching to or adding either IL‐6 inhibitor or IL‐1 inhibitor is recommended.	1	C	100.0%	9.9 (0.5)
8.	In patients with no biologic DMARD access OR resistant to both IL‐1 inhibitors and IL‐6 inhibitors, addition of or switching to any of the following medications may be considered. Calcineurin inhibitors (cyclosporin A or tacrolimus), particularly in patients with systemic features Methotrexate, particularly in patients with predominant articular manifestation Thalidomide Cyclophosphamide JAK inhibitors All these strategies require great caution on toxicity monitoring.	2	D (CsA) C (MTX) D (THA, CTX, JAKi)	100.0%	9.6 (0.7)
9.	There is insufficient evidence to recommend for or against IVIG, colchicine, or etoposide.	—	D (IVIG) N (Col, Eto)	100.0%	9.8 (0.5)
The following recommendation statements apply to children and young adults with Still's disease without MAS who are in clinically inactive disease
10.	We suggest tapering off systemic glucocorticoids as soon as inactive disease is achieved, with the goal of discontinuation ideally within 6 months of glucocorticoid initiation.	2	C	100.0%	10.0 (0)
11.	We suggest maintaining a sustained inactive disease for at least 6 months after systemic glucocorticoid discontinuation, prior to tapering DMARDs, regardless of DMARD classes initiated.	2	D	87.5%	9.2 (1.1)
12.	We suggest extending the dosing interval of biologic DMARDs rather than decreasing the dosage.	2	D	100.0%	9.5 (0.9)
13.	In patients with the combination of conventional synthetic DMARD and biologic DMARD, tapering either of the medications first may be considered. There is no preferred order of tapering.	2	D	100.0%	9.4 (0.8)
The following recommendation statements apply to children and young adults with Macrophage Activation Syndrome associated with active Still's disease
14.	If there is no contraindication, we recommend pulse methylprednisolone as initial therapy as soon as possible.	1	B	100.0%	10.0 (0.2)
15.	In patients with inadequate response to pulse methylprednisolone within the first 48–72 h, we suggest the addition of cyclosporin A or anakinra or IFN‐γ inhibitors (i.e., emapalumab). IL‐6 inhibitors may be considered with caution.	2	C (CsA, ANK, IFN‐γi) D (TCZ)	100.0%	9.6 (0.7)
16.	Plasma exchange or etoposide might be considered in patients who are refractory to biologic DMARDs and cyclosporin A.	2	D (PLEX) C (Eto)	93.8%	9.2 (0.9)
17.	JAK inhibitors might be considered in patients who are refractory to biologic DMARDs and conventional synthetic DMARDs.	2	D (BAR, RUX, TOF)	100.0%	9.4 (0.9)
18.	In order to achieve the ultimate goal (drug‐free remission), the following intermediate targets are suggested: At day 7, resolution of fever and reduction of CRP by > 50%, At week 4, no fever, reduction of active (or swollen) joint count by > 50%, normal CRP and physician and patient/parent global assessment less than 20 on a 0–100 VAS, At month 3, CID with glucocorticoids less than 0.2 mg/kg/day, At month 6, CID without glucocorticoids.	2	D	100.0%	9.6 (0.7)

Note: SoR (Strength of Recommendation): 0 = absence of SoR, 1 = Strong Recommendation, 2 = Weak Recommendation; CoE (Certainty of Evidence): A = High, B = Moderate, C = Low and D = Very low; %A (Agreement): Percentage of consensus panel member agreeing with the recommendation statement.

Abbreviations: ANK, anakinra; BAR, baricitinib; Col, colchicine; CsA, cyclosporin A; DMARD, disease modifying anti‐rheumatic drug; Eto, etoposide; IFN‐γi, interferon γ inhibitor; IVIG, intravenous immunoglobulin; LoA, Level of Agreement (0 totally disagree −10 totally agree; mean [SD, standard deviation]); MAS, macrophage activation syndrome; N, no evidence; PLEX, plasma exchange; RTX, rituximab; RUX, ruxolitinib; TCZ, tocilizumab; TNFi, tumor necrosis factor inhibitor; TOF, Tofacitinib.

Early recognition and prompt initiation of therapy constitute the cornerstone of JIA management (OP1). This paradigm is particularly salient in Still's disease, where timely intervention has consistently demonstrated superior efficacy and improves long‐term outcomes, most notably with biologic agents, such as IL‐1i and IL‐6i (window of opportunity) [20, 21, 22]. Accurately diagnosing Still's disease can be challenging, particularly in the absence of arthritis. Malignancies, infectious diseases, autoinflammatory and other immune‐mediated inflammatory diseases must be carefully excluded, providing that the infectious diseases are more prevalent in the region and the use of high‐dose glucocorticoids can mask other immune‐mediated or inflammatory disease manifestations [23]. The diagnostic approach to Still's disease is not addressed in this recommendation; practitioners are advised to consult existing guidelines [13, 23].

Management of patients with Still's disease needs to be adapted to the individual, taking into account the different circumstances (cost, drug availability and insurance practices), values and preferences, health beliefs, and cultural and ethnic background of each patient. This approach increases adherence and therefore improves outcomes. Treatment target and treatment strategy should be discussed through shared decision‐making between the patient, caregiver, and treatment team (OP2).

Treat‐to‐target refers to a target‐oriented strategy for the management of chronic diseases, as opposed to the traditional symptom‐focused methodology. This involves frequent treatment adjustment, guided by quantitative indices and has been shown to improve outcomes in adult rheumatoid arthritis irrespective of the therapeutic agent used [24, 25, 26]. A similar effect was seen in a recent German JIA multicenter cohort study (BiKER cohort) [27]. The targets are divided into short‐, medium‐ and long‐term targets (see below) due to the urgent need to control the inflammation in order to avoid life‐threatening complications, including MAS and the window of opportunity for the treatment of Still's disease (OP3) [20, 21, 22].

While the sequelae of glucocorticoid administration are well‐characterized, certain pediatric complications—including cataracts, dermal striae, and avascular necrosis—represent significant, permanent morbidities. In resource‐limited regions such as the Asia Pacific, access to biologic disease modifying anti‐rheumatic drugs (bDMARDs)—particularly IL‐6i, and even more so, IL‐1i—is constrained, whether due to prohibitive costs or outright unavailability [11]. Consequently, glucocorticoids remain indispensable for achieving rapid control of systemic inflammation in Still's disease. Moreover, in MAS complicating Still's disease, high‐dose pulse methylprednisolone continues to represent the first‐line therapeutic choice in our region. Nonetheless, the TF advocates systemic glucocorticoids be administered at the lowest effective dose for the shortest possible duration (OP4).

3.2. Consensus Statements on Still's Disease Management (Table 1)

3.2.1. Induction Therapy

Treatment of naïve, newly diagnosed active Still's disease without MAS (Figure 2)

1. We suggest against NSAIDs as initial monotherapy (PICO 1)

FIGURE 2.

Treatment algorithm for Still's disease without macrophage activation syndrome. Treatment algorithms should always be interpreted in conjunction with individual clinical judgment, taking into account each patient's unique circumstances. bDMARD, biological disease‐modifying anti‐rheumatic drug; CID, clinically inactive disease; CNI, calcineurin inhibitors; CsA, cyclosporin A; GC, glucocorticoid; IL‐1i, interleukin‐1 inhibitors; IL‐6i, interleukin‐6 inhibitors; JAKi, Jak kinase inhibitors; MAS, macrophage activation syndrome; MTX, methotrexate.

Non‐steroidal anti‐inflammatory drugs (NSAIDs) are weakly recommended as initial monotherapy in the American College of Rheumatology (ACR) recommendations [12]. However, they are mentioned in the newer European Alliance of Associations for Rheumatology (EULAR)/Pediatric Rheumatology European Society (PReS) recommendations to use for symptomatic treatment of fever and pain during the diagnostic workup period [13]. Additional observational studies, including one from the region, demonstrated that the majority of patients initiated on NSAIDs monotherapy required additional glucocorticoids or other disease modifying anti‐rheumatic drugs (DMARDs) [28, 29]. Given the limited availability of healthcare facilities, the scarcity of pediatric rheumatologists, and the low disease awareness across many areas in the region, the TF advises against the use of NSAIDs as first‐line monotherapy in order to avoid further delays in initiating appropriate treatment. This RS is a weak recommendation based on very low CoE.

• 2 We suggest against systemic glucocorticoids as monotherapy. However, systemic glucocorticoids may be used as a bridging therapy. (PICO 1)

According to the ACR recommendations, oral glucocorticoids are weakly recommended against as initial monotherapy [12]. They are instead primarily indicated as interim therapy while awaiting access to bDMARDs or as bridging treatment until conventional synthetic DMARDs (csDMARDs) achieve therapeutic effect, particularly within the first 3 months. Glucocorticoids as bridging therapy were defined as a short course of oral form (≤ 2 mg/kg/day with a maximum of 60 mg/day). The current recommendations are consistent with those of the ACR. However, the SoR is weak, reflecting the very low CoE. Observational studies have indicated that early initiation of glucocorticoids may accelerate remission, though they do not alter the overall disease trajectory [30, 31, 32, 33]. Nonetheless, the TF recognizes that in circumstances where access to biologics is not feasible, the combination of glucocorticoids with csDMARDs is necessary to achieve control of systemic and/or articular disease. As outlined in OP4, glucocorticoids should be administered at the lowest effective dose for the shortest required duration to minimize long‐term adverse complications.

• 3 If possible, we suggest initiating treatment with IL‐6 inhibitors or IL‐1 inhibitors. However, conventional synthetic DMARDs may be considered if IL‐6 inhibitors or IL‐1 inhibitors are not available. (PICO 2, 3)

Both Western guidelines recommend initiating bDMARDs targeting IL‐1 or IL‐6 as monotherapy and emphasize that treatment should commence as soon as possible [12, 13]. In the absence of controlled studies, no preferred agent was endorsed. Recent meta‐analyzes demonstrate that both IL‐1 and IL‐6 inhibitors are efficacious, significantly improving ACR50 response rates at week 4 compared to baseline. Pooled analysis of four randomized controlled trials (RCTs) evaluating three IL‐1i (rilonacept, anakinra, canakinumab) demonstrated a significant association with treatment response (OR = 6.92, 95% CI 2.24–21.36) [34]. Similarly, tocilizumab treatment was associated with an increased likelihood of response compared with placebo (OR = 8.08, 95% CI: 1.89–34.57) [34]. Real‐world longitudinal studies also demonstrated that early initiation of IL‐1i or IL‐6i was associated with high rates of clinically inactive disease [34]. In addition, five recent observation studies, comparing first‐line or early bDMARDs with those who did not, reported that early initiation of biologics, either IL‐1i or IL‐6i, was linked to higher rates of event‐free remission and successful glucocorticoid withdrawal compared with csDMARDs [35, 36, 37, 38, 39]. Recent evidence supports the efficacy and safety of biologic therapy in Still's disease. Tocilizumab (TCZ), evaluated in two open‐label and eight observational studies, demonstrated high efficacy with minimal serious adverse events [40, 41, 42, 43, 44, 45, 46, 47, 48, 49]. Anakinra (ANK), across one RCT and two observational studies, showed consistently high response rate [50, 51, 52]. Similarly, canakinumab (CAM), assessed in two open‐label and five real‐world observational studies, reinforces its role not only in refractory but also earlier disease, with substantial proportions achieving early response. Efficacy was sustained during long‐term follow‐up without significant adverse events [53, 54, 55, 56, 57, 58, 59, 60]. Furthermore, a recent network meta‐analysis of RCTs on the efficacy and safety of bDMARDs in Still's disease. Nine trials with 430 patients were included. All bDMARDs were associated with greater odds of ACR50 response compared with placebo. There was no statistically significant association between biologic drugs and serious adverse events. The multivariate meta‐analysis found no difference between bDMARDs (ANK, CAM, TCZ and rilonacept) [61]. Accordingly, no sequence for bDMARD initiation is currently recommended.

Findings from our recent regional survey indicated that while TCZ is available in most of the participating countries, access to ANK or CAM is considerably more limited, with fewer than half of the countries reporting availability compared to the widespread accessibility of csDMARDs. Despite some degree of access to these bDMARDs, their high cost continues to represent a major barrier to utilization [11]. Consequently, csDMARDs remain the predominant therapeutic option across much of the Asia‐Pacific region. To prevent therapy delay, the TF suggests initiation of csDMARDs in combination with a bridging glucocorticoids regimen to control active disease when bDMARDs are unavailable/feasible. This constitutes a weak recommendation despite the moderate overall CoE for early bDMARD, reflecting limitations in regional resource availability.

• 4 When conventional synthetic DMARDs are being considered as initial therapy, methotrexate may be considered for patients with predominant articular disease, while cyclosporin may be considered for those with predominant systemic disease. In some cases, a combination of the two may also be considered. This strategy requires careful toxicity monitoring. (PICO 2)

The 2021 ACR guideline strongly recommends against csDMARDs as initial monotherapy despite very low CoE, whereas the 2024 EULAR/PReS recommendation does not address csDMARDs as first‐line therapy [12, 13]. Both Western guidelines, however, acknowledged the role of csDMARDs in circumstances where IL‐1i or IL‐6i are inaccessible or unavailable. As outlined above, both bDMARD classes demonstrate superior efficacy compared with csDMARDs monotherapy, achieving a higher rate of inactive disease, sustaining effect during long‐term follow‐up, and reducing glucocorticoids exposure to discontinuation, all with minimal serious adverse events including MAS observed [34, 61].

Recent systematic literature review (SLR) indicates that csDMARDs are employed either following inadequate glucocorticoid response or in combination with bDMARDs [34]. Evidence regarding methotrexate (MTX) in Still's disease is limited by heterogeneity in dosage and administration routes, complicating cross‐study comparisons [62, 63, 64, 65, 66]. Nonetheless, MTX has demonstrated efficacy, achieving inactive disease in up to 70% of patients, and significantly reducing glucocorticoid dose, irrespective of concomitant arthritis [63, 64, 65, 67].

Observational studies of calcineurin inhibitors (CNI), including cyclosporine (CsA) and tacrolimus, have demonstrated efficacy in Still's disease, with longer event‐free survival, higher rates of clinical remission, and reduced glucocorticoid requirements compared to patients without CNI [68, 69, 70]. MTX was also combined with CsA in one prospective and two retrospective studies [71, 72, 73], yielding clinically inactive disease (CID) rates ranging from 73% to 86%.

Thalidomide has been investigated in two observational studies—one prospective and one retrospective, demonstrating disease control rates of 73%–85%, with improvements in both arthritis and fever [74, 75]. Reported adverse events included transient paresthesia, reversible aminotransferase elevations, and somnolence. Owing to its teratogenic potential and the likelihood that these recommendations will be applied by non‐rheumatologists, the TF advises caution and does not advocate thalidomide use except under exceptional circumstances.

The TF recommends MTX for predominantly articular disease and CsA for predominantly systemic disease, with the option of MTX + CsA combination therapy in severe or refractory cases. Given the potential for increased adverse events, this approach should be applied with caution. This recommendation is based on regional expert consensus and supported by evidence of low to very low certainty, resulting in a weak SoR.

• 5 We recommend against using TNF inhibitors as initial therapy. (PICO 3)

No trials of TNF inhibitors (TNFi) have been conducted in Still's disease. However, a recent SLR and meta‐analysis reported pooled efficacy rates (ACR70 or CID) of 23% (95% CI 13%–34%) in AOSD and 28% (95% CI 15%–43%) in sJIA [34]. The TF upgrades the strength of this recommendation despite very low CoE, reflecting both the comparatively limited efficacy of Tumor necrosis factor‐inhibitors (TNFi) relative to other DMARDs and the consensus panel's prioritization of avoiding harm from treatment delays. Consequently, the SoR is high.

The following recommendation statements apply to children and young adults with Still's disease without MAS who do not respond to initial therapy.

• 6 In patients initiated on an IL‐6 inhibitor or IL‐1 inhibitor at their respective therapeutic dose (or after dose optimization), switching to either IL‐1 inhibitor or IL‐6 inhibitor is recommended, if feasible. (PICO 5)

Evidence from randomized trials and observational studies demonstrates that IL‐1i or IL‐6i provide higher efficacy, faster remission, and lower toxicity compared with csDMARDs. Recent data further support their role in facilitating glucocorticoid discontinuation, in contrast to conventional agents, and underscore the risks associated with long‐term glucocorticoid exposure [34]. Switching between IL‐1i and IL‐6i has been shown to yield comparable outcomes, consistent with the 2021 ACR guideline [12, 53, 58, 76, 77]. On this basis, the TF upgraded the recommendation. Accordingly, the SoR is strong despite the low CoE, reflecting the clinical benefits of bDMARDs and the emphasis on minimizing glucocorticoid use. This also highlights that patients may respond differently to IL‐1i and IL‐6i, allowing treatment flexibility and avoiding therapeutic delay.

• 7 In patients initiated with a conventional synthetic DMARD at a therapeutic dose (or after dose optimization), if possible, switching to or adding either IL‐6 inhibitor or IL‐1 inhibitor is recommended. (PICO 5)

Randomized controlled trials and observational studies consistently demonstrate the superior efficacy and safety of IL‐1 and IL‐6 inhibitors in Still's disease [34]. The initiation of csDMARDs should be reserved for circumstances in which both bDMARD classes are inaccessible or infeasible. Two scenarios warrant specific consideration: (1) if either IL‐1i or IL‐6i subsequently become available, switching to or adding one of these agents is recommended, as current evidence does not indicate superiority of one over the other; and (2) if both biologic options remain unavailable, a different class of csDMARD—with or without bridging glucocorticoids—should be initiated promptly to control active disease. This recommendation is strengthened despite low CoE, reflecting the clear therapeutic benefit of IL‐1 and IL‐6 inhibitors and the imperative to avoid delays in disease control.

• 8
  In patients with no biologic DMARD access or resistant to both IL‐1 inhibitors and IL‐6 inhibitors, addition of or switching to any of the following medications may be considered.

  • Calcineurin inhibitors (cyclosporin A or tacrolimus), particularly in patients with systemic features.
  • Methotrexate, particularly in patients with predominant articular manifestation.
  • Thalidomide
  • Cyclophosphamide
  • JAK inhibitors

All these strategies require great caution on toxicity monitoring. (PICO 2, 4).

Supporting evidence for the use of CNI, MTX, and thalidomide has been discussed above. Regarding cyclophosphamide, two observational studies have evaluated its use in sJIA. The first involved 18 patients treated with a regimen of three consecutive days of pulse methylprednisolone, low‐dose intravenous cyclophosphamide (400 mg/m²) on the third day, and oral MTX (10 mg/m²/week), repeated every 3 months [67]. The second study included four patients who received 6–10 monthly intravenous pulse methylprednisolone (30 mg/kg, max 1 g) in combination with cyclophosphamide (500–1000 mg/m²) [78]. Both reports suggested some therapeutic efficacy; however, the CoE remains very low.

For JAK inhibitors (JAKi), five small case series involving 7–14 patients with sJIA or AOSD, together with one meta‐analysis, have reported outcomes with tofacitinib, baricitinib, ruxolitinib, or upadacitinib [79, 80, 81, 82, 83, 84]. Complete remission was achieved in up to 71% of patients, while partial remission was observed in up to 43%. The role of JAKi in Still's disease without MAS remains confined to refractory cases that have failed IL‐1 and IL‐6 inhibitors, with or without concomitant csDMARDs.

This statement is a weak recommendation based on low to very low certainty of supporting evidence.

• 9 There is insufficient evidence to recommend for or against intravenous immunoglobulin (IVIG), colchicine, or etoposide. (PICO 4, 6)

Evidence from one small RCT and four observational studies with variable follow‐up did not demonstrate a significant effect of IVIG on prognosis or overall outcomes in Still's disease, although steroid‐sparing benefits were noted [34, 85, 86]. Colchicine was primarily used for Still's disease–associated serositis in a single study [34], while no evidence supports the use of etoposide outside of MAS. Accordingly, the TF considers the available data insufficient to recommend these agents.

3.2.2. Maintenance Therapy

The following recommendation statements apply to children and young adults with Still's disease without MAS who are in clinically inactive disease.

• 10 We suggest tapering off systemic glucocorticoids as soon as inactive disease is achieved, with the goal of discontinuation ideally within 6 months of glucocorticoid initiation. (PICO 7)

The TF acknowledges the necessity of high‐dose glucocorticoids (> 1 mg/kg/day) to control initial disease activity as bridging therapy, particularly in settings where bDMARDs are unavailable. The long‐term adverse effects of glucocorticoids are well established, notably growth impairment in children [87]. In such circumstances, a treatment duration of at least 2–3 months is often unavoidable; however, once therapeutic targets are achieved, tapering should be undertaken progressively, with discontinuation ideally within 6 months—a goal the TF considers attainable. The TF recognizes that a proportion of patients may require ongoing glucocorticoid therapy at variable doses, often in combination with a csDMARD or combined‐csDMARDs to achieve steroid‐sparing effects. Nevertheless, every effort should be directed toward minimizing glucocorticoid exposure by using the lowest effective dose for the shortest possible duration.

• 11 We suggest maintaining a sustained inactive disease for at least 6 months after systemic glucocorticoid discontinuation, prior to tapering DMARDs, regardless of DMARD classes initiated. (PICO 7, 8)

Data guiding the management of DMARDs following achievement of CID remain limited. The TF recommends discontinuation of glucocorticoids as the initial mandatory step. Although no definitive trial has established the optimal timing for DMARD tapering, the PRINTO MTX withdrawal study demonstrated that maintaining CID for 6 or 12 months did not alter flare rates; notably, sJIA was included despite small numbers [88, 89]. Regression analyzes adjusted for ILAR subtype revealed no differential risk of flare across subtypes. Based on these findings, the TF advises maintaining CID for at least 6 months off glucocorticoids before initiating DMARD tapering. A stepwise reduction every 3–6 months is recommended, consistent with prior Western guidelines, allowing re‐escalation if flare occurs [13]. This approach aims to mitigate severe relapse risk and harmonize DMARD withdrawal strategies across JIA subtypes.

• 12 We suggest extending the dosing interval of biologic DMARDs rather than decreasing the dosage. (PICO 8)

Evidence is lacking to guide DMARD tapering following achievement of CID. The TF suggests extending the dosing interval of bDMARDs to minimize drug wastage, align with patient values and preferences, and enhance adherence by reducing injection frequency.

• 13 In patients with the combination of conventional synthetic DMARD and biologic DMARD, tapering either of the medications first may be considered. There is no preferred order of tapering. (PICO 8)

This is a weak recommendation, supported by very low‐CoE from a single study. A recent observational analysis found no significant difference in flare rates whether MTX or bDMARDs were withdrawn first [90].

3.2.3. Macrophage Activation Syndrome (Figure 3)

FIGURE 3.

Treatment algorithm for Still's disease with macrophage activation syndrome. Treatment algorithms should always be interpreted in conjunction with individual clinical judgment, taking into account each patient's unique circumstances. ASAP, as soon as possible; CsA, cyclosporin A; IFNγi, interferon‐γ inhibitor; IL‐1i, interleukin‐1 inhibitors; MAS, macrophage activation syndrome; PLEX, plasma exchange.

The following recommendation statements apply to children and young adults with Macrophage Activation Syndrome associated with active Still's disease.

• 14 If there is no contraindication, we recommend pulse methylprednisolone as initial therapy as soon as possible. (PICO 9)

Macrophage activation syndrome (MAS) is a life‐threatening complication of Still's disease. Identification of triggering factors is essential, with infection being more common in the Asia‐Pacific region; this necessitates prompt and thorough investigation, followed by targeted antimicrobial therapy. Such measures facilitate the management of both MAS and the underlying Still's disease. High‐dose glucocorticoids remain the mainstay therapy and should be initiated without delay, typically with pulse methylprednisolone at 15–30 mg/kg/day (maximum 1 g/dose) for 3–5 days. Dexamethasone may be considered as an alternative, particularly in cases with central nervous system involvement, given its superior penetration of the blood–brain barrier.

• 15 In patients with inadequate response to pulse methylprednisolone within the first 48–72 h, we suggest the addition of cyclosporin A or IL‐1 inhibitors or IFN‐γ inhibitors (i.e., emapalumab). IL‐6 inhibitors may be considered with caution. (PICO 9)

High‐dose glucocorticoid therapy has demonstrated satisfactory efficacy in a substantial subset of patients with MAS associated with Still's disease, particularly when commenced early in the disease course. In severe or refractory MAS, where high‐dose glucocorticoids prove inadequate, timely initiation of additional therapy within 48–72 h is recommended. To date, no RCTs have addressed MAS in Still's disease, apart from a recent small open‐label study of emapalumab [91]. The TF highlights substantial and favorable evidence supporting CNIs, particularly CsA (oral or intravenous), in combination with glucocorticoids for disease control and steroid‐sparing effects, as demonstrated in a SLR and a recent observational report [34, 68]. In resource‐limited regions such as the Asia Pacific, the TF suggests the addition of a CNI following failure of high‐dose glucocorticoid therapy, given their wider availability and accessibility compared with IL‐1i or emapalumab [11].

A recent meta‐analysis demonstrated that MAS could occur during treatment with IL‐1 or IL‐6 inhibitors, with pooled incidence rates of 2.7/100 patient‐years (95% CI 1.8–3.9) for TCZ and 2.2/100 patient‐years (95% CI 2.4–3.5) for ANK. Practitioners should exercise caution when MAS develops in patients with Still's disease receiving these agents, as key parameters—including fever and inflammatory markers—may be attenuated [92]. Observational studies of ANK in sJIA‐associated MAS (n = 44) reported a complete response in 73% of patients, though concomitant CsA use was not clearly defined [34]. Emapalumab, the first monoclonal antibody targeting IFN‐γ, demonstrated efficacy and safety in an open‐label trial of patients with MAS refractory to high‐dose glucocorticoids. Administered at an initial dose of 6 mg/kg followed by 3 mg/kg every 3 days until day 15, then twice weekly until day 28, in combination with high‐dose methylprednisolone, emapalumab induced remission in 93% of patients by week 8, with the earliest response at day 9 [91]. However, its high cost and limited availability preclude widespread use in the Asia‐Pacific region. By contrast, TCZ is more accessible, and recent regional observational studies have reported favorable efficacy and safety in MAS [93, 94, 95]. Given its higher incidence of MAS and greater masking of fever and inflammatory markers, particularly CRP, IL‐6 inhibition is generally reserved as second‐line therapy after failure of other agents. In resource‐limited settings, the TF anticipates CsA as first‐line therapy following glucocorticoid failure, with TCZ positioned subsequently in the treatment sequence.

• 16 Plasma exchange or etoposide might be considered in patients who are refractory to biologic DMARDs and cyclosporin A. (PICO 10)

Plasma exchange, typically employed in combination with other therapies, has been reported in treatment‐refractory Still's disease–associated MAS, predominantly in small observational studies, with favorable outcomes when therapeutic options are otherwise exhausted [34, 96, 97]. Similarly, low‐dose etoposide has demonstrated efficacy across several observational cohorts, generally administered alongside other agents, with evidence of improved survival [34, 98].

This weak recommendation reflects the very low CoE supporting plasma exchange and low‐dose etoposide in refractory MAS associated with Still's disease. The available data are limited to small observational studies, with inherent methodological constraints and risk of bias. Nevertheless, the TF acknowledges that in many parts of the region, therapeutic options are restricted, and patients with multiple drug failures face high mortality risk. In such contexts, consideration of these modalities may be warranted, as their use has been associated with improved survival in observational cohorts. The balance of potential benefit against the uncertainty of evidence underpins the conditional nature of this recommendation.

• 17 JAK inhibitors might be considered in patients who are refractory to biologic DMARDs and conventional synthetic DMARDs. (PICO 9)

Although evidence is limited to small observational studies, JAK inhibitors—particularly the JAK1/JAK2 agents ruxolitinib and baricitinib—have shown efficacy in controlling MAS complicating Still's disease and in improving survival [34, 80, 81, 82, 84, 99, 100]. Reported outcomes indicate clinically meaningful benefit in patients with otherwise treatment‐refractory MAS. In view of the paucity of effective alternatives and the high associated mortality, the TF considers JAKi a potential therapeutic option while underscoring the conditional/weak nature of this recommendation given the very low CoE.

• 18
  In order to achieve the ultimate goal (drug‐free remission), the following intermediate targets are suggested:

  • At day 7, resolution of fever and reduction of CRP by > 50%,
  • At week 4, no fever, reduction of active (or swollen) joint count by > 50%, normal CRP and physician and patient/parent global assessment less than 20 on a 0–100 VAS,
  • At Month 3, CID with glucocorticoids less than 0.2 mg/kg/day,
  • At Month 6, CID without glucocorticoids.

In line with the overarching principle of treat‐to‐target (T2T), where treatment is modified according to the disease activity. This approach is particularly encouraged in Still's disease management, where delayed treatment may result in severe or life‐threatening complications. Consistent with the immediate targets (within 6 months) outlined in the EULAR/PReS recommendations, the TF adopts these targets, recognizing their simplicity and feasibility in guiding treatment across the region [34]. These targets are intended as guidance; however, in severely resource‐limited settings, complete glucocorticoid discontinuation may not be achievable. In such circumstances, glucocorticoids may be maintained at the lowest effective dose for the shortest possible duration, while considering change or escalation of DMARD.

4. Discussion

These consensus recommendations represent the second part of the APLAR JIA Guidelines Task Force initiative, following the regional consensus on polyarticular course JIA (pJIA), temporomandibular joint arthritis, and non‐pharmacological management of JIA (in press). In contrast to pJIA, Still's disease is a clinically distinct autoinflammatory syndrome driven predominantly by IL‐1 and IL‐6 pathways [13]. The therapeutic landscape for Still's disease has evolved rapidly over the past decade; however, access to biologic agents and implementation of treat‐to‐target strategies remain highly variable across the Asia‐Pacific region. This variability is compounded by a critical shortage of pediatric rheumatologists, with most patients managed by non‐pediatric rheumatologist practitioners, underscoring the need for a tailored regional approach. While conceptually aligned with recent international guidelines, including the 2021 ACR recommendations [12] and the 2024 EULAR/PReS consensus recommendations [13], the APLAR statements introduce region‐specific modifications. These adaptations reflect both emerging evidence and pragmatic considerations unique to the Asia‐Pacific context (Table 2).

TABLE 2.

Differences and similarities of the current APLAR recommendation with the existing Western guidelines.

Domain	APLAR guideline	2024 EULAR/PReS	2021 ACR
Treat to target (T2T)	Strongly frames management as T2T with the ultimate goal of drug‐free remission; defines targets at day 7, week 4, month 3, month 6		Does not specify a time‐anchored T2T target
Treatment naïve Still's disease without MAS: NSAIDs	Suggests against NSAIDs as initial monotherapy	The algorithm prioritizes early IL‐1/IL‐6 inhibitors without mentioning NSAIDs as a core initial monotherapy strategy	Conditionally recommends NSAIDs as initial monotherapy
Treatment naïve Still's disease without MAS: glucocorticoids	Suggests against systemic GCs as monotherapy, but allow bridging	Uses GCs by severity; emphasizes tapering as soon as possible with the explicit target of CID low dose at 3 months and CID off GCs at 6 months.	Conditionally recommends against oral GCs as initial monotherapy, but acknowledges use when biologics are not available; in the lowest dose/shortest duration possible
Treatment naïve Still's disease without MAS: IL1/IL‐6 inhibitors	Suggests initiating IL‐1/IL‐6 inhibitors if possible; however, allows fallback to csDMARDs if biologics are not available.	Emphasizes early initiation with IL‐1/IL‐6 inhibitors once the diagnosis is established	Conditionally recommends IL‐1/IL‐6 inhibitors as initial monotherapy without preferred agent
Treatment naïve Still's disease without MAS: csDMARDs	Considers the use of MTX for predominant arthritis; CsA for systemic disease when biologics not available	Not positioned as first‐line. The management algorithm is a biologic‐first approach	Strongly recommends against csDMARDs as initial monotherapy in Still's disease without MAS
Treatment naïve Still's disease without MAS: TNFi	Specifically recommends against TNFi as initial therapy.	Did not explicitly recommend against the use of TNFi as initial therapy, but emphasizes the use of IL‐1/6 inhibitors	TNFi not recommended as initial Still's disease therapy
Still's disease without MAS with inadequate response to initial IL‐1/IL‐6 inhibitor strategy	Supports switching to another class when initial therapy fails.
Refractory disease/no biologic access/resistant to IL‐1 + IL‐6 inhibitors	Lists options of CNI, MTX, thalidomide, cyclophosphamide and JAK inhibitors	Does not mention alternatives to IL‐1/IL‐6 inhibitors if not available	Notes options for residual arthritis after IL‐1/IL‐6 inhibitors (e.g., add MTX, switch to abatacept or TNFi)
Inactive disease: glucocorticoids	All support tapering off GCs but APLAR and EULAR/PReS gives specific target timepoints for GC discontinuation
Inactive disease: tapering biologics	Maintains sustained inactive disease ≥ 6 months after GC discontinuation before tapering DMARDs	Maintenance of CID 3–6 months off GCs before bDMARD tapering; and ≥ 6 months CID off GCs before tapering is reiterated in the algorithm narrative	Conditionally recommends tapering/discontinuing bDMARDs after inactive disease, but timing/method unclear due to limited evidence
Inactive disease: tapering strategy	Prefers extend dosing interval rather than decrease dosage	Not specifically mentioned	Not specifically mentioned
MAS: initial therapy	Recommends pulse methylprednisolone ASAP if no contraindication.		Conditionally recommends GCs as part of initial treatment for Still's disease with MAS
MAS: escalation of therapy	If inadequate response within 48–72 h, suggests adding CsA or IL‐1 inhibitors or IFN‐γ inhibitors. Mentions the use of IL‐6 inhibitors with caution	Only considers IL1 inhibitors instead of IL‐6 inhibitors. Also mentions cyclosporin A and/or IFN‐γ inhibitor as part of initial therapy	Conditionally recommends IL‐1/IL‐6 inhibitors over CNI alone to achieve inactive disease and MAS resolution; acknowledges combination therapy may be needed in severe MAS.
MAS: refractory disease	PLEX or etoposide may be considered if refractory. JAK inhibitors may be considered	Not specifically mentioned	Formal recommendations deferred
Complications	Not present in the excerpted APLAR statements.	Has explicit statements: active screening (symptoms + PFTs; HRCT if symptomatic)	Discusses concern about Still's disease–associated lung disease

Abbreviations: CID, clinically inactive disease; CNI, calcineurin inhibitor; CsA, cyclosporin A; csDMARD, conventional synthetic disease modifying anti‐rheumatic drug; GCs, glucocorticoids; HRCT, high resolution computed tomography; JAK, janus kinase; MAS, macrophage activation syndrome; PFT, pulmonary function test; PLEX, plasma exchange; RTX, rituximab; RUX, ruxolitinib; TCZ, tocilizumab; TNFi, tumor necrosis factor inhibitor; TOF, Tofacitinib.

4.1. Early Therapeutic Strategy and Discouragement of NSAID or Glucocorticoid Monotherapy

A major distinction of the APLAR consensus recommendations is the explicit discouragement of NSAIDs and systemic glucocorticoids as initial monotherapy for treatment‐naïve, active Still's disease without MAS, with preference given to early initiation of IL‐1 or IL‐6 inhibitors where feasible. This contrasts with the 2021 ACR guideline, which conditionally recommended NSAIDs as possible first‐line therapy in selected patients [12] but aligns with the 2024 EULAR/PReS recommendations emphasizing early biologic intervention [13]. Two considerations underpin this divergence. First, the APLAR task force recognizes Still's disease as a cytokine‐driven autoinflammatory condition mediated primarily by IL‐1 and IL‐6. Observational data demonstrate that NSAID monotherapy rarely achieves sustained remission and may delay timely escalation; more than half of patients initiated on NSAIDs ultimately require glucocorticoids or DMARDs [28, 29]. Second, glucocorticoid exposure remains a major determinant of long‐term morbidity, particularly growth impairment in children [87]. Accordingly, the TF chose to include a steroid‐sparing strategy as one of the explicit therapeutic objectives.

This recommendation reflects a more aggressive “window‐of‐opportunity” approach to Still's disease, aiming to prevent chronic joint damage and reduce steroid dependency through early cytokine blockade. Nevertheless, access to biologic agents remains highly variable across the Asia‐Pacific region. To accommodate differences in regulatory approval and drug availability, the consensus statements deliberately frame biologic therapy as “if possible,” ensuring flexibility while emphasizing the importance of early intervention where feasible.

4.2. Role of Conventional Synthetic DMARDs

In contrast to the 2021 ACR guidelines, which did not endorse csDMARD monotherapy for Still's disease [12], the current APLAR consensus adopts a more pragmatic stance. Although IL‐1 or IL‐6 inhibitors remain the preferred first‐line therapy, MTX or CNI may be considered when biologics are unavailable. A greater consideration is given to csDMARD in this guideline to reflect real‐world practice in many lower‐ and middle‐income countries across the Asia‐Pacific region. The recommendations are further stratified by phenotype: MTX may provide benefit in patients with predominant articular involvement, whereas CNI may be more suitable for predominant systemic inflammation and MAS phenotypes, despite the low CoE. Inclusion of csDMARDs thus balances limited evidence with the practical realities of regional healthcare systems and aligns with global efforts to expand access to essential therapies for childhood rheumatic diseases. Notably, MTX is listed in the pediatric section of the WHO Model List of Essential Medicines [101], underscoring its importance as a core treatment option in settings where biologics are inaccessible.

4.3. Explicit Rejection of TNF Inhibitors as Initial Therapy

The APLAR consensus explicitly recommends against the use of TNFi as initial therapy for Still's disease. While prior guidelines did not endorse TNFi as first‐line biologics, the current statement provides clearer guidance, supported by mechanistic and clinical evidence indicating that TNF blockade fails to adequately control systemic inflammation in this condition [34]. Although TNFi may be more accessible in the Asia‐Pacific region due to biosimilars and lower cost, their use risks delaying initiation of IL‐1 or IL‐6 inhibitors, or other potentially more effective csDMARDs. This explicit recommendation is intended to prevent inappropriate therapeutic sequencing and to optimize timely initiation of effective therapy.

4.4. Treat‐To‐Target Approach and Drug Tapering

Similar to the 2024 EULAR/PReS guideline, the APLAR consensus statement outlines treatment targets at Day 7, Week 4, Month 3, and Month 6, culminating in glucocorticoid‐free inactive disease by month 6 [13]. These milestones were not defined in the 2021 ACR guideline [12]. This reflects the adoption of a treat‐to‐target approach, consistent with paradigms established in other inflammatory rheumatic diseases [102]. The consensus statement specifies quantitative targets—including fever resolution, reduction in inflammatory markers, improvement in joint counts, and, critically, defined glucocorticoid thresholds—to reduce therapeutic inertia and encourage timely escalation. This framework also aims to harmonize outcome monitoring across diverse healthcare systems. However, these targets have not yet been validated in the Asia‐Pacific setting, where access to bDMARDs, particularly IL‐1 and in some contexts IL‐6 inhibitors, remains limited. Whether such targets can be consistently achieved under current resource constraints is therefore uncertain and constitutes a priority for future research. Modifications toward more attainable targets may be required following regional practice experience and data collection.

Another distinct point in this consensus statement is the tapering strategy for patients who achieve CID on biologic therapy. This has not been explored in previous guidelines. Our consensus statement suggests extending the dosing interval rather than reducing the individual dose. The evidence base for drug tapering in Still's disease is sparse. With improved outcomes and more patients achieving CID, practical tapering strategies are necessary. While interval extension is a more feasible approach, this recommendation should be interpreted cautiously, as it is based on very low CoE and requires validation in prospective studies.

4.5. Expanded Framework for Refractory Disease

For patients refractory to both IL‐1 and IL‐6 inhibitors, or without access to biologics, the APLAR consensus provides a broader list of potential agents, including CNI, thalidomide, cyclophosphamide, and JAKi. Evidence supporting these therapies remains of very low certainty, yet their inclusion reflects real‐world practice in parts of Asia. The conditional consideration of thalidomide is particularly contextual: although not widely endorsed in Western guidelines due to toxicity concerns, structured monitoring systems in some centers have enabled its use with demonstrable efficacy [74, 75]. Rather than omitting such options, the task force elected to provide explicit cautionary guidance to ensure clarity and safe application in resource‐constrained environments. The monitoring should include electrophysical evaluation with electromyography (EMG) and nerve conduction velocity (NCV) tests, if feasible.

4.6. Macrophage Activation Syndrome Management and Escalation Pathways

In cases of MAS complicating Still's disease, the APLAR consensus recommends pulse methylprednisolone as initial therapy, with subsequent addition of CsA, IL‐1i, particularly ANK, or emapalumab (IFN‐γ inhibitor) in the event of inadequate response. This approach is consistent with prior guidelines, which recognized IL‐1 blockade and IFN‐γ inhibition as effective strategies for MAS, though access to these agents remains limited across most of the Asia‐Pacific countries [12, 13]. For refractory disease, the consensus panel also acknowledges potential roles for etoposide, JAKi, and plasma exchange. Inclusion of these options underscores both the severity of refractory MAS and the need for flexible escalation pathways, particularly in settings where availability of targeted therapies is constrained.

4.7. Addressing Gaps in Evidence

Despite advances in understanding the pathophysiology of Still's disease, high‐quality evidence remains scarce across most PICO domains. Observational studies are frequently underpowered due to small sample sizes, and outcome definitions remain heterogeneous. There is a pressing need for adequately designed RCTs evaluating therapies beyond IL‐1 and IL‐6 inhibitors. Although biologics are increasingly regarded as first‐line therapy in Still's disease without MAS, the relative efficacy of IL‐1 versus IL‐6 blockade remains uncertain. With improving outcomes and growing numbers of children achieving sustained remission, further data are required to define the optimal duration of biologic therapy prior to tapering. In addition, as experience with targeted synthetic DMARDs expands, cost‐effectiveness analyzes will become increasingly important, particularly in low‐ and middle‐income countries in the Asia‐Pacific region. Addressing these knowledge gaps will necessitate regional collaboration and the establishment of patient registries.

Recent policy developments have significantly advanced global access to therapies for childhood rheumatic diseases. The WHO Model List of Essential Medicines (EML) has been updated to include agents for pediatric inflammatory joint disease; however, the current list is limited to MTX, intra‐articular glucocorticoids, and selected TNFi [101]. While the EML establishes a minimum standard for medicines that should be available within national health systems, key cytokine‐targeted therapies for Still's disease, such as IL‐1 and IL‐6 inhibitors, remain excluded. This disparity highlights the gap between emerging evidence and evolving global policy frameworks. Sustained advocacy will be required to ensure that essential medicine lists keep pace with emerging evidence in pediatric rheumatology and reflect the therapeutic needs of children worldwide.

Several limitations must be acknowledged in applying this consensus statement to clinical practice. First, although the GRADE methodology was systematically applied, many recommendations rely on indirect evidence, particularly regarding tapering strategies and management of refractory MAS. Second, heterogeneity in study populations—including variability in inclusion of childhood‐ versus adult‐onset Still's disease—and inconsistency in defining “inactive disease” limit comparability and reduce applicability to specific patient groups. Third, economic considerations are highly relevant in Asia‐Pacific settings, yet formal studies remain scarce. While pragmatic and real‐world factors influenced the development of these recommendations, they were not formally incorporated into the evidence synthesis. Finally, restricting the literature search to English‐language publications introduces the potential for language and publication bias and may have resulted in the omission of relevant studies from the Asia‐Pacific region. This is particularly pertinent given that clinical experience and real‐world evidence on Still's disease may be reported in local languages, including Japanese and Chinese, potentially limiting the representation of regional practice patterns, uncommon complications such as macrophage activation syndrome, and treatment strategies influenced by local healthcare systems and resource availability. However, the English‐language restriction was adopted to ensure a feasible, consistent, and methodologically rigorous review process across participating review teams. To mitigate this limitation, the TF comprises experts from across the Asia‐Pacific region who interpreted the evidence in the context of regional clinical practice and healthcare constraints. Furthermore, the review deliberately included csDMARDs alongside biologic therapies to reflect treatments that remain widely available in many countries within the region. These measures were intended to enhance the applicability of the recommendations despite the limitations of the published evidence.

5. Conclusion

This APLAR consensus represents a distinct and necessary evolution of existing guidance for Still's disease by explicitly addressing the realities of healthcare delivery across the Asia‐Pacific region. While international recommendations have advanced a biologic‐driven, treat‐to‐target paradigm supported by biologics, these strategies assume timely access to specialist care and high‐cost therapies. In contrast, the APLAR recommendations are deliberately contextualized to heterogeneous health systems in which access to biologic agents remains variable. Importantly, this work does not merely adapt existing recommendations but reframes them through a resource‐stratified lens, integrating evidence with real‐world feasibility and health system constraints. Persistent gaps in high‐quality evidence highlight the need for collaborative research and regional registries to refine management strategies for this rare but potentially severe disease.

Author Contributions

T.A., S.F., L.F.D. and K.L.T. conceptualized the study. All authors contributed to drafting, discussing, modifying, and finalizing the consensus statements. T.A. and K.L.T. drafted the manuscript. K.L.T., C.M.A., K.A., P.C., M.C.M.C.‐C., N.G.U.C., M.T.C., C.H.J., S.C.L., M.J.H.L., B.L., W.S. and M.B.V. did the systematic literature search and review. All authors edited and approved the final manuscript.

Funding

The current APLAR Consensus Recommendation development was funded by APLAR SIG Grant 2023. APLAR had no influence on evidence appraisal, recommendation wording, voting procedures, or manuscript preparation.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

R.A.J. reports honoraria from Pfizer. All other authors declare no competing interests. This work was funded by the APLAR secretariat to support the online voting platform and other secretary supports. There are no commercial conflicts of interest to be declared.

Supporting information

Appendix S1: Task force members' list and their responsibility.

Appendix S2: Evidence report and Summary of Finding (SoF) tables.

Appendix S3: Search Inclusion Criteria.

Appendix S4: Search Strategies.

Figure S1: Preferred Reporting Items for Systematic reviews and Meta‐Analyzes (PRISMA) Flowchart.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.