Rheumatology 2021;60:2421–2426

Rheumatology doi:10.1093/rheumatology/keaa634
Advance Access publication 17 November 2020

Original article

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
Clinical significance of interleukin-18 for the
diagnosis and prediction of disease course in
systemic juvenile idiopathic arthritis
Mao Mizuta1, Masaki Shimizu 1,a
, Natsumi Inoue1, Yasuhiro Ikawa1,
Yasuo Nakagishi , Ryuhei Yasuoka3, Naomi Iwata3 and Akihiro Yachie1
2

Abstract
Objective. To investigate the clinical significance of serum IL-18 levels for the diagnosis of systemic JIA (s-JIA)
and to predict the disease course of s-JIA.
Methods. Overall, 116 patients with s-JIA, 151 with other diseases and 20 healthy controls were analysed. Serum
IL-18 levels were measured longitudinally in 41 patients with s-JIA from active phase through remission phase.
Serum IL-18 levels were quantified via enzyme-linked immunosorbent assay and the results were compared with
clinical features and the disease course of s-JIA.
Results. The serum IL-18 level cut-off value for differentiation of s-JIA from other diseases was 4800 pg/ml. In
patients with a monocyclic course, serum IL-18 levels steadily decreased during the inactive phase and low levels
were sustained during remission. In contrast, in patients with a chronic course, elevated serum IL-18 levels were
sustained even during the inactive phase. In patients with a polycyclic course, serum IL-18 levels were elevated
during disease flares and normalized during the inactive phase. The serum IL-18 level cut-off value for diagnosis of
remission in s-JIA was 595 pg/ml,
Conclusion. Serum IL-18 levels of >4800 pg/ml may be useful for differentiating between s-JIA and other dis-
eases. Monitoring of serum IL-18 levels might be useful for predicting the disease course and assessing remission
in s-JIA.
Key words: interleukin-18, systemic juvenile idiopathic arthritis, diagnosis, remission

CL IN IC A L
SC I E NC E
Rheumatology key messages

. A serum IL-18 cut-off level of >4800 pg/ml was useful for differentiating between s-JIA and other diseases.
. A serum IL-18 cut-off level of <595 pg/ml was useful for the diagnosis of remission in s-JIA.
. Monitoring of serum IL-18 levels might be useful in the prediction of disease course.

Introduction approximately half of patients have a monocyclic

course, experiencing remission within 2–4 years [2],
Systemic JIA (s-JIA) is a unique subtype of JIA, charac- while others have a relapsing course. Some show a
terized by arthritis and other systemic features such as polycyclic course with repeated disease flares featuring
spiking fever, salmon-coloured skin rash, hepatospleno- systemic symptoms and/or arthritis [2]. Most patients,
megaly, generalized lymphadenopathy, and polyserositis though, have a chronic course with chronic arthritis,
[1]. The disease course of s-JIA is highly variable; which is usually more prominent after regression of

Correspondence to: Masaki Shimizu, Department of Pediatrics,

1
Department of Pediatrics, School of Medicine, Institute of Medical, Graduate School of Medical Sciences, Kanazawa University, 13-1
Pharmaceutical, and Health Sciences, Kanazawa University, Takaramachi, Kanazawa 920-8641, Japan.
Kanazawa, 2Department of Pediatric Rheumatology, Hyogo E-mail: [email protected]
Prefectural Kobe Children’s Hospital, Kobe and 3Department of a
Present address: Department of Child Health and Development,
Infection and Immunology, Aichi Children’s Health and Medical Graduate School of Medical and Dental Sciences, Tokyo Medical and
Center, Obu, Japan Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510,
Submitted 7 July 2020; accepted 1 September 2020 Japan

C The Author(s) 2020. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: [email protected]
V
Mao Mizuta et al.

systemic features and typically lasts no longer than TABLE 1 Clinical characteristics of 116 patients with sys-
5 years [2]. temic JIA
Recent studies have shown that s-JIA might be driven

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
by innate proinflammatory cytokines, with IL-1, IL-6 and Patients
IL-18 playing important roles in pathogenesis [3]. (N 5 116)
Furthermore, biologic therapies that block IL-1 and IL-6
are seen to have dramatic effects in patients with s-JIA Characteristics
[4–6]. We previously reported increased serum IL-18 lev- Age at disease onset: 5 (0.9–16)
median (range), years
els in active s-JIA [7, 8]. Serum IL-18 levels are even fur-
Sex: male/female, n 61/55
ther increased in patients with s-JIA–related
Duration: median (range), days 12 (1–90)
macrophage activation syndrome; this increase seems Clinical symptoms
significant, considering other forms of secondary hae- Fever, n 116
mophagocytic lymphohistiocytosis (HLH) are associated Rash, n 93
with lower serum IL-18 levels [7]. Serum IL-18 levels are Hepatomegaly, n 15
therefore increasingly used as a biomarker for s-JIA Splenomegaly, n 11
diagnosis. However, it remains unclear whether serum Lymphadenopathy, n 39
IL-18 level can effectively differentiate s-JIA from other Pleuritis, n 3
Pericarditis, n 8
inflammatory diseases and, if so, what the cut-off value
Affected joint counts: 1 (0–30)
should be. median (range), n
Previous reports have also shown that serum IL-18 Laboratory findings
levels may reflect s-JIA disease activity [7–11]. Our pre- Ferritin: median (range), ng/ml 1189 (63–57 010)
liminary data showed serum IL-18 levels in patients with WBC: median (range), /ml 14 900 (4050–38 200)
s-JIA with a good disease course decreased to CRP: median (range), mg/dl 9.39 (0.18–32.97)
<1000 pg/ml in the inactive phase, compared with sus- AST: median (range), IU/l 33 (8–1071)
tained elevated levels in patients with a chronic course LDH: median (range), IU/l 342 (70–1480)
[9]. It is likewise still unclear whether serum IL-18 levels
are closely correlated with s-JIA disease course, and WBC: white blood cell; AST: asparartate transaminase;
are useful as a diagnostic laboratory criterion for LDH: lactate dehydrogenase.
remission.
In this study, we aimed to investigate the clinical sig- onset of s-JIA, with the presence of arthritis being con-
nificance of serum IL-18 levels for diagnosis of s-JIA, firmed later. Sera from patients with other diseases
particularly to determine the cut-off value of serum IL-18 were obtained during the active phase of disease. One
levels to differentiate s-JIA from inflammatory diseases. patient with TRAPS was treated with a low dose of
To do this, we compared serum IL-18 levels in patients prednisone. No treatments, such as prednisone, colchi-
with s-JIA with those in patients with inflammatory dis- cine, immunosuppressants or biologics were used for
eases. Furthermore, we aimed to investigate the clinical any of the other patients.
significance of serum IL-18 levels for predicting s-JIA Serum IL-18 levels were measured longitudinally in 41
disease course and assessing remission. s-JIA patients in the active phase, including disease
onset and disease flare through inactive phase or remis-
sion. The median follow-up term of this study was
Methods 709 days. Forty patients were treated with prednisolone
(PSL). Twenty-five patients were treated with ciclosporin,
Patients and samples 3 with tacrolimus, 1 with MTX and 1 with mizoribine.
Thirty patients were treated with tocilizumab (TCZ). The
In total, 116 subjects with s-JIA, 78 with Kawasaki dis-
median duration from disease onset to commencement
ease (KD), 7 with FMF with a mutation in exon 10 of
of TCZ was 2 months. The clinical characteristics of
MEFV (5 patients with M694I, 2 with M694V), 3 with TNF
these 41 patients with s-JIA are shown in Table 2.
receptor–associated periodic syndrome (TRAPS), 23
The criteria for s-JIA in the active phase were as fol-
with other subtypes of JIA (10 oligoarticular, 10 polyar-
lows: active arthritis, fever, rash, splenomegaly, general-
ticular and 3 enthesitis-related arthritis), 10 with SLE, 12
ized lymphadenopathy, serositis, and elevation of serum
with JDM, 6 with leukaemia (4 acute lymphoblastic leu-
CRP level and/or ESR. The criteria for s-JIA in the in-
kaemia, 1 acute myelogenous leukaemia and 1 juvenile
active phase were as follows: no arthritis, no fever, no
myelomonocytic leukaemia), and 20 age-matched
rash, no splenomegaly, no generalized lymphadenop-
healthy controls (HCs) were enrolled in the study. All s-
athy, no serositis, normal serum CRP level and ESR,
JIA patients were newly diagnosed, and their sera were
and Physician Global Assessment of disease activity
obtained before commencing treatment with any im-
indicating no disease activity. Remission criteria were
munosuppressant, steroid or biologic agents. Clinical
based on guidelines proposed by Wallace et al. [13].
characteristics of the 116 s-JIA patients are shown in
The disease course of s-JIA was defined as follows
Table 1. Diagnosis of s-JIA was based on the ILAR crite-
[2]: (1) monocyclic course—no disease flare; (2) chronic
ria [12]. Some patients had minimal joint disease at

2422 https://academic.oup.com/rheumatology
 The significance of IL-18 in systemic JIA

TABLE 2 Clinical characteristics of s-JIA patients longitu- FIG. 1 Serum IL-18 levels in patients with s-JIA and
dinally measured serum IL-18 levels from active phase to other inflammatory diseases
remission

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
Patients
(N 5 41)

Characteristics
Age at disease onset: median (range), years 4 (0.67–15)
Sex: male/female, n 16/25
Treatment
PSL, n 40
CsA, n 25
Others, n 5
TCZ, n 30
Clinical course
Monocyclic course 20
Chronic course 11
Polycyclic course 10

TCZ: tocilizumab; PSL: prednisolone.

course—disease flare occurring frequently after steroid

withdrawal; (3) polycyclic pattern—disease flare occur- s-JIA; systemic JIA; KD; Kawasaki disease; TRAPS; TNF
ring repeatedly after achieving remission. receptor-associated periodic syndrome; HCs; healthy
Sera were extracted from blood samples, divided into controls. Bars represent median values. Statistically sig-
aliquots, frozen, and stored at 80 C until analysis. This nificant differences between patient groups are shown
study was approved by the institutional review board of as follows, *P < 0.05, **P < 0.01, ***P < 0.001, ****P <
Kanazawa University, and all specimens were used after 0.0001.
the receipt of informed patient consent.
range 48–455 pg/ml; P < 0.0001), SLE (median 825 pg/
Measurement of serum IL-18 levels ml, range 225–1800 pg/ml; P < 0.01), JDM (median
438 pg/ml, range 86–1250 pg/ml; P < 0.0001), leukaemia
Serum IL-18 levels were measured using ELISA accord-
(median 826 pg/ml, range 207–1480 pg/ml; P < 0.05),
ing to the manufacturer’s instructions (IL-18: MBL,
and HCs (median 133 pg/ml, range 76–215 pg/ml; P <
Nagoya, Japan).
0.0001). Serum IL-18 levels in patients with s-JIA were
not significantly elevated compared with patients with
Statistical analysis
FMF (median 3050 pg/ml, range 1180–4400 pg/ml), al-
Statistical analyses were performed using GraphPad though this difference was statistically significant when
Prism 7 software (GraphPad, San Diego, CA, USA). Data comparing between the two groups using Mann–
were expressed as median and range. Comparisons be- Whitney’s test (P < 0.05).
tween several groups were performed using one-way
analysis of variance with Tukey’s multiple comparison
test. Comparisons between two groups were made Serum IL-18 level cut-off value for differentiation of
using the Mann–Whitney U test. In all cases, P < 0.05 s-JIA from other inflammatory diseases
was considered statistically significant. As shown in Table 3, receiver operating characteristic
(ROC) curve analysis revealed serum IL-18 level cut-off
values for differentiation of s-JIA from KD, FMF, TRAPS,
Results other subtypes of JIA, SLE, JDM, leukaemia and HCs of
4560 pg/ml, 4800 pg/ml, 1685 pg/ml, 1728 pg/ml, 2400 pg/
Serum IL-18 levels were significantly elevated in
ml, 2125 pg/ml, 2240 pg/ml and 1608 pg/ml, respectively.
patients with s-JIA compared with other inflamma-
ROC area under the curve (AUC) values were 0.993,
tory diseases
0.9938, 1.0, 1.0, 1.0, 1.0, 1.0 and 1.0, respectively.
As shown in Fig. 1, serum IL-18 levels in patients with s- Sensitivities were 94.9%, 100%, 100%, 100%, 100%,
JIA (median 34 750 pg/ml, range 3000–34 0000 pg/ml) 100%, 100% and 100%, respectively. Specificities were
were significantly elevated compared with patients with 99.1%, 99.1%, 100%, 100%, 100%, 100%, 100% and
KD (median 360 pg/ml, range 60–12 800 pg/ml; P < 100%, respectively.
0.0001), TRAPS (median 275 pg/ml, range 180–370 pg/ The serum IL-18 level cut-off value for differentiation
ml; P < 0.05), other subtypes of JIA (median 130 pg/ml, of s-JIA from all other diseases was 4800 pg/ml. The

https://academic.oup.com/rheumatology 2423
Mao Mizuta et al.

TABLE 3 ROC curve analysis of serum IL-18 levels for the s-JIA from other inflammatory diseases with similar clin-
differentiation s-JIA from other diseases ical manifestations such as KD, other autoinflammatory

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
diseases, other rheumatic diseases, including SLE,
Cut-off value (pg/ml) AUC JDM, and other subtypes of JIA, and leukaemia.
Furthermore, some patients with s-JIA had minimal joint
KD 4560 0.993 disease at disease onset, with the presence of arthritis
FMF 4800 0.9938 being confirmed later. Therefore, s-JIA is among the
TRAPS 1685 1.0 most important underlying causes of fever of unknown
Other subtypes of JIA 1728 1.0 origin. In a clinical setting, there is a growing need for a
SLE 2400 1.0
promising new biomarker that is specific for the diagno-
JDM 2125 1.0
Leukaemia 2240 1.0 sis of s-JIA.
All of other diseases 4800 0.9958 In this study, serum IL-18 levels were significantly ele-
HCs 1608 1.0 vated in s-JIA compared with in other inflammatory dis-
eases. The serum IL-18 level cut-off value for
KD: Kawasaki disease; TRAPS: TNF-associated periodic differentiation of s-JIA from other diseases was 4800 pg/
syndrome; HCs: healthy controls. ml, with a sensitivity of 97.1% and a specificity of
99.1%. Serum IL-18 levels of >4800 pg/ml might there-
fore be useful for diagnosing s-JIA. However, serum IL-
ROC AUC value was 0.9958 (Table 3), sensitivity was 18 levels also increase in XIAP deficiency [14] and
97.1% and specificity was 99.1%. NLRC4 mutation-associated recurrent MAS with early-
onset enterocolitis [15]. These diseases should therefore
Serum IL-18 levels predict disease course of s-JIA
be carefully differentiated from s-JIA.
Next, we investigated whether longitudinal monitoring In this study, we demonstrated that serum IL-18 level
of serum IL-18 levels could predict disease course. In reflected disease course of s-JIA. In patients with mono-
s-JIA patients with a monocyclic course, serum IL-18 phasic course, serum IL-18 levels were less than ap-
levels steadily decreased in the inactive phase and low proximately 1000 pg/ml in the inactive phase and
levels were sustained during remission (Fig. 2A). In s-JIA normalized during remission. On the other hand, in
patients with a chronic course, elevated serum IL-18 patients with chronic course, elevated serum IL-18 lev-
levels of >1000 pg/ml were sustained even in the in- els of >1000 pg/ml were sustained even in the inactive
active phase (Fig. 2B). In s-JIA patients with a polycyclic phase. In patients with a polycyclic pattern, serum IL-18
course, serum IL-18 levels were elevated to >1000 pg/ levels were elevated to >1000 pg/ml during flares, but
ml during disease flares and decreased to <1000 pg/ml normalized again in the inactive phase. Serum IL-18 lev-
in the inactive phase (Fig. 2C). els therefore change in agreement with the disease
course. From these findings, monitoring of serum IL-18
Serum IL-18 level cut-off value for diagnosis of level might be useful for predicting the disease course
remission in s-JIA of s-JIA.
As shown in Fig. 3, serum IL-18 levels in patients with s- In the clinical setting of s-JIA, flares have been seen
JIA were significantly elevated in the active phase com- to occur during the inactive phase [7]. Previous studies
pared with in the inactive phase. Furthermore, serum IL- showed circulating M2-type monocytes tend to increase
18 levels in patients with s-JIA were significantly ele- as a proportion of monocytes in clinically inactive dis-
vated in the inactive phase compared with in the remis- ease [16]. Furthermore, serum levels of HO-1, sCD163
sion phase. ROC curve analysis revealed a serum IL-18 and IL-10 remain elevated, whereas clinical parameters
level cut-off value for differentiation of the inactive phase and other pro-inflammatory cytokines normalize [17].
from the active phase was 6975 pg/ml. The ROC AUC These findings for clinically inactive disease reflect a
value was 0.9171, the sensitivity was 88.4%, and the state of compensated inflammation in s-JIA. In this
specificity was 84.8%. The serum IL-18 level cut-off study, longitudinal examination of serum IL-18 levels in
value for differentiation of the remission phase from the patients with a chronic disease course revealed sus-
inactive phase was 595 pg/ml. The ROC AUC value was tained elevation during the inactive phase. These find-
0.7752, the sensitivity was 69.7% and the specificity ings indicate that persistent inflammation associated
was 79.1%. As shown in Supplementary Fig. S1, avail- with macrophage activation was not being controlled.
able at Rheumatology online, no statistically significant Therefore, patients with elevated serum IL-18 levels dur-
differences were observed in serum IL-18 levels be- ing inactive and remission phases should be carefully
tween patients who did and did not receive tocilizumab. monitored when treatment is withdrawn.
In this study, serum IL-18 levels were significantly ele-
vated in the inactive phase compared with in the remis-
sion phase, during which they were normalized. These
Discussion
findings support the accuracy of the Wallace criteria for
Diagnosis of s-JIA is based mainly on clinical features. clinical remission [13]. The serum IL-18 level cut-off val-
Therefore, it is sometimes challenging to differentiate ues for the differentiation of remission phase from

2424 https://academic.oup.com/rheumatology
 The significance of IL-18 in systemic JIA

FIG. 2 Longitudinal examination of serum IL-18 levels in patients with systemic JIA

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
(A) Monocyclic course, no disease flare; (B) chronic course, disease flare occurring frequently after steroid withdrawal;
(C) polycyclic course, disease flare occurring repeatedly after achieving remission. Red arrows show timing of disease
flares in each patient.

FIG. 3 Serum IL-18 levels in each disease phase in between patients with s-JIA who did or did not receive
patients with systemic JIA TCZ. These findings indicate that serum IL-18 levels
might not be affected by IL-6 blocking. Therefore,
monitoring of serum IL-18 levels is also useful for
those patients with s-JIA treated with TCZ.
Interestingly, serum IL-18 levels were normal in all
patients with s-JIA who did not receive TCZ, whereas
levels in some patients with s-JIA who received TCZ
were elevated to >1000 pg/ml even in the inactive and
remission phases. Most of these patients showed a
chronic disease course. TCZ can mask clinical symp-
toms of s-JIA; therefore, patients with s-JIA receiving
TCZ with an elevated serum IL-18 level during the in-
active and remission phases should be carefully moni-
tored when treatment is withdrawn.
The limitation of this study is that it did not include
patients with s-JIA treated with IL-1 antagonists such as
anakinra and canakinumab. In Japan, canakinumab has
recently been approved for the treatment of s-JIA (from
July 2018), while anakinra has not yet been approved.
Serum IL-18 levels in patients with systemic JIA in each Further studies including patients treated with IL-1
disease phase are shown. Bars represent median val- antagonists may help to define the true diagnostic value
ues. Statistically significant differences between patient of IL-18.
groups are shown as follows, *P < 0.05, ****P < 0.0001. In conclusion, we demonstrated that serum IL-18 lev-
els of >4800 pg/ml might be useful for the differenti-
ation of s-JIA from other diseases. Serum IL-18 levels
inactive phase was 595 pg/ml; therefore, levels of reflect disease activity and predict disease course of s-
<595 pg/ml may be a useful criterion for clinical remis- JIA. Furthermore, serum IL-18 levels of <595 pg/ml
sion in s-JIA. may provide a useful diagnostic laboratory criterion for
Recent studies have shown that biologics can mod- clinical remission in s-JIA. Thus, monitoring of serum
ify clinical symptoms and laboratory data such as IL-18 levels may prove useful for the assessment
serum CRP levels [18, 19]. In this study, no significant of disease activity and prediction of disease course in
differences in serum IL-18 levels were observed s-JIA.

https://academic.oup.com/rheumatology 2425
Mao Mizuta et al.

Acknowledgements with particular emphasis on the role of interleukin-18 in

its pathogenesis. Rheumatology 2010;49:1645–53.
We thank Ms Harumi Matsukawa and Ms Mizue Mizuno

Downloaded from https://academic.oup.com/rheumatology/article/60/5/2421/5983858 by Biblioteca IRCCS Ospedale Pediatrico Bambino Gesu - Roma user on 25 October 2023
8 Shimizu M, Nakagishi Y, Inoue N et al. Interleukin-18 for
for technical assistance. predicting the development of macrophage activation
Funding: This work was supported by the Japan Society syndrome in systemic juvenile idiopathic arthritis. Clin
Immunol 2015;160:277–81.
for the Promotion of Science (JAPS) KAKENHI
(18K07786). 9 Shimizu M, Nakagishi Y, Yoshida A, Yachie A. Serum
interleukin 18 as a diagnostic remission criterion in
Disclosure statement: The authors have declared no systemic juvenile idiopathic arthritis. J Rheumatol 2014;
conflicts of interest. 41:2328–30.
10 Jelusic M, Lukic IK, Tambic-Bukovac L et al. Interleukin-
18 as a mediator of systemic juvenile idiopathic arthritis.
Data availability statement Clin Rheumatol 2007;26:1332–4.
All data generated or analysed during this study are 11 Shigemura T, Yamazaki T, Hara Y et al. Monitoring
included in this published article (and its supplementary serum IL-18 levels is useful for treatment of a patient
information files). with systemic juvenile idiopathic arthritis complicated by
macrophage activation syndrome. Pediatr Rheumatol
Online J 2011;9:15.
Supplementary data 12 Petty RE, Southwood TR, Manners P et al. International
League of Associations for Rheumatology classification
Supplementary data are available at Rheumatology
of juvenile idiopathic arthritis: second revision,
online.
Edmonton. J Rheumatol 2004;31:390–2.
13 Wallace CA, Giannini EH, Huang B, Itert L, Ruperto N,
References Childhood Arthritis Rheumatology Research Alliance
(CARRA). American College of Rheumatology provisional
1 Ravelli A, Grom AA, Behrens EM, Cron RQ. Macrophage criteria for defining clinical inactive disease in select
activation syndrome as part of systemic juvenile categories of juvenile idiopathic arthritis. Arthritis Care
idiopathic arthritis: diagnosis, genetics, pathophysiology Res 2011;63:929–36.
and treatment. Genes Immun 2012;13:289–98.
14 Wada T, Kanegane H, Ohta K et al. Sustained elevation
2 Singh-Grewal D, Schneider R, Bayer N, Feldman BM. of serum interleukin-18 and its association with hemo-
Predictors of disease course and remission in systemic phagocytic lymphohistiocytosis in XIAP deficiency.
juvenile idiopathic arthritis: significance of early clinical Cytokine 2014;65:74–8.
and laboratory features. Arthritis Rheum 2006;54:
15 Canna SW, de Jesus AA, Gouni S et al. An activating
1595–601.
NLRC4 inflammasome mutation causes
3 Mellins ED, Macaubas C, Grom AA. Pathogenesis of autoinflammation with recurrent macrophage activation
systemic juvenile idiopathic arthritis: some answers, syndrome. Nat Genet 2014;46:1140–6.
more questions. Nat Rev Rheumatol 2011;7:416–26.
16 Macaubas C, Nguyen KD, Peck A et al. Alternative
4 Yokota S, Imagawa T, Mori M et al. Efficacy and safety activation in systemic juvenile idiopathic arthritis
of tocilizumab in patients with systemic-onset juvenile monocytes. Clin Immunol 2012;142:362–72.
idiopathic arthritis: a randomised, double-blind, placebo-
17 Shimizu M, Yachie A. Compensated inflammation
controlled, withdrawal phase III trial. Lancet 2008;371:
in systemic juvenile idiopathic arthritis: role of alternatively
998–1006.
activated macrophages. Cytokine 2012;60:226–32.
5 Verbsky JW, White AJ. Effective use of the recombinant
18 Shimizu M, Nakagishi Y, Kasai K et al. Tocilizumab
interleukin 1 receptor antagonist anakinra in therapy
masks the clinical symptoms of systemic juvenile
resistant systemic onset juvenile rheumatoid arthritis.
idiopathic arthritis-associated macrophage activation
J Rheumatol 2004;31:2071–5.
syndrome: the diagnostic significance of interleukin-18
6 Ruperto N, Brunner HI, Quartier P et al. Two randomized and interleukin-6. Cytokine 2012;58:287–94.
trials of canakinumab in systemic juvenile idiopathic
19 Schulert GS, Minoia F, Bohnsack J et al. Effect of
arthritis. N Engl J Med 2012;367:2396–406.
biologic therapy on clinical and laboratory features of
7 Shimizu M, Yokoyama T, Yamada K et al. Distinct macrophage activation syndrome associated with
cytokine profiles of systemic-onset juvenile idiopathic systemic juvenile idiopathic arthritis. Arthritis Care Res
arthritis-associated macrophage activation syndrome 2018;70:409–19.

2426 https://academic.oup.com/rheumatology