Systemic juvenile idiopathic arthritis (sJIA) is a systemic inflammatory disease clinically characterized by fever, lymphadenopathy, arthritis, rash, and serositis. sJIA is the most serious subtype of juvenile idiopathic arthritis (1) and accounts for 30% to 40% of all JIA in Asia (2). A significant number of patients develop severe disease and treatment-related complications such as persistent arthritis, growth delay, and osteoporosis. Serious complications which are potentially fatal including macrophage activation syndrome (MAS) occur in 10% to 15% of children with sJIA (3–5). sJIA is divided into three phenotypes: monocyclic, polycyclic, and persistent course (6, 7).

The etiology of sJIA is not fully understood; proinflammatory cytokines including interleukin (IL)-6, IL-1, and IL-18 play an important role in the pathogenesis of the disease. IL-6 mediates systemic inflammation in sJIA, leading to joint synovial hyperplasia and joint destruction (8–10). Blockade of IL-6 represents the main mechanism of sJIA treatment and prevention of potential complications (11).

In 2011, the United States and Europe successively approved tocilizumab, a humanized monoclonal antibody TCZ against the IL-6 receptor for treating children with sJIA. Due to the heterogeneity of sJIA, the clinical response of patients treated with TCZ is different; hence, it is needed to better characterize the profile of patients with sJIA who are more likely to respond to IL-6 blockade. Pacharapakornpong et al. (12) found that in the early TCZ treatment (<6 months), sJIA patients had a higher remission rate than late TCZ treatment (>6 months). In this study, we performed a single-center retrospective study to determine the different efficacy of TCZ in the early and late stages and in three phenotypic subgroups (monocyclic, polycyclic, and persistent) of sJIA. The safety profile and therapeutic effect of TCZ were observed and analyzed to provide a clinical basis for the treatment of children with sJIA.

This study represented 67.5 years of TCZ exposure in 77 sJIA patients; adverse events occurred in 21 patients (Table 3). Leukopenia was observed in seven patients, including one leukopenia induced by streptococcal infection, two infusion reactions characterized by fever and cold chills, and facial blushing, which occurred during the second infusion of TCZ. This was relieved by discontinuing the infusion and intravenous dexamethasone treatment and did not reoccur during later TCZ infusions. One infusion reaction occurred during the sixth TCZ infusion, presented as fever, chills, and cyanosis, and was also relieved by discontinuing the infusion and intravenous dexamethasone treatment. For this patient, TCZ treatment was terminated. Two patients experienced MAS, one at 3 months and the other at 6 months of tocilizumab treatment. Two patients had continuous active disease during tocilizumab treatment, and MAS was improved after comprehensive treatment instead of TCZ.

The treatment of sJIA usually requires NSAIDs, DMARDs, and glucocorticoids (6). It is necessary to use biological agents when it is refractory to glucocorticoids and DMARDs (10, 16). The pathophysiological basis of sJIA is the activation of pro-inflammatory cytokines, especially IL-1β and IL-6 (17, 18). Currently, there are two main biological treatment strategies for sJIA: IL-1 and IL-6 biologic blockade (19, 20). However, in China, IL-1 blockers are still unavailable so that IL-6 blockers are the main biologic treatment for sJIA.

In our study, we found that fever, active arthritis, sJADAS-27 score, white blood cell count, ESR, and CRP can be relieved after 2 weeks of TCZ treatment. TCZ has significant short-term efficacy for sJIA. Furthermore, we observed sJIA patients with TCZ administration in early stage and found that it had better remission of active arthritis than those administrated in the late stage after 12 weeks of treatment. TCZ can quickly relieve fever and inflammation (indicated by decreased CRP and ESR) (21–24), especially for patients showing less active arthritis in the early stage. Doaa et al. (25) identified that younger patients with shorter disease duration and greater systemic manifestations showed more favorable outcomes by TCZ therapy. These observations are consistent with the window of opportunity hypothesis, which suggests that IL-6 blockade may be more effective in early sJIA, when the disease is characterized by more prominent systemic presentation and less active arthritis (26). Likewise, Alexeeva et al. (27) found that only earlier age at initiation of TCZ therapy was a statistically significant factor associated with reaching the best response to therapy in polyarticular JIA.

The course of sJIA varies and is divided into three phenotypes (7). In this study, sJIA was classified as monocyclic (n = 12, 15.6%), polycyclic (n = 14, 18.2%), and persistent course (n = 51, 66.2%). Bielak et al. (28) defined three phenotypes of sJIA: monocyclic, polycyclic, and polyarticular disease, which occurs as arthritis flares in >4 joints. Systemic inflammation, such as fever, elevated CRP, and ESR, is not more prominent in patients with multi-joint sJIA than single-joint sJIA, and systemic inflammation gradually evolves into an autoimmune disease phenotype (24, 29). The heterogeneity of sJIA contributes to the differences. Bielak et al. found that polycyclic and monocyclic sJIA responded better to tocilizumab than polyarticular sJIA. In our study, the clinical response was worst in patients with persistent course and the best in monocyclic. A previous study indicated that IL-1 inhibitors may be useful if patients do not respond to TCZ. sJIA with persistent course often show polyarticular JIA, and in these patients, TCZ is effective for fever and inflammation, but not for polyarthritis. Therefore, sJIA with persistent course who do not respond to TCZ may need an IL-1 antagonist or a TNF-a monoclonal antibody (30). To obtain optimal therapeutic responses, it is necessary to predict the disease phenotype of sJIA at an early stage. Singh-Grewal et al. (7) proposed that the clinical features observed at 3 months (the presence of active arthritis and fever) and 6 months (elevated ESR > 26 mm/H requiring corticosteroid treatment) are accurate predictors of a patient's disease course. This information helps identify patient's risk of developing into persistent disease and having a higher likelihood of a poor functional outcome, which need timely therapeutic intervention to prevent from joint damage and disability. Therefore, early TCZ treatment is recommended for such patients with early predictions that may be persistent courses.

The most common adverse effect observed in our study was granulocytopenia, which was mild and not accompanied by severe infection. Clinical remission was achieved in six of the seven cases, suggesting that patients who develop granulocytopenia after TCZ treatment may have a greater chance to obtain clinical remission. Neutropenia associated with sJIA is dependent on IL-6 levels, and leukocytopenia and granulocytopenia may be used as biomarkers of susceptibility to treatment with IL-6 monoclonal antibodies (31). There were two cases of MAS, which was not unexpected given that previous studies report that 20% to 25% of patients with sJIA treated with biological agents develop MAS (32). Current research suggests that TCZ does not prevent the occurrence of MAS, and indeed in some patients, TCZ may even trigger this reaction. The specific cause and mechanism are unclear. It is possible that when a patient experiences an obvious episode of active sJIA, TCZ antagonizes IL-6 which could trigger a negative feedback loop, therefore causing amplified and excessive inflammation and inducing MAS. Therefore, the timing of TCZ treatment is a critical factor.

In this study, glucocorticoids were standardized to ≤1 mg/kg/day to avoid the effect of high doses, but almost 60%–70% patients enrolled were taken at baseline. In addition, we did not study the beneficial effects of TCZ on glucocorticoid reduction and catch-up growth. Moreover, due to economic conditions, TCZ was not added every 2 weeks strictly, which could have a partial impact on the research results. It was a single-center retrospective study with low number of patients enrolled, so prospective cohort studies and further multicenter clinical studies are needed to shed additional light on this matter and ultimately improve the care of patients with sJIA.

We demonstrated that TCZ can quickly relieve fever and inflammation, especially when patients have less active arthritis with shorter disease duration. The long-term efficacy of TCZ is related to the phenotype; therefore, it is necessary to predict the disease phenotype of sJIA at an early stage. These findings may help to define the profile of patients with sJIA who are more likely to benefit from TCZ.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Written informed consent was obtained from the minor(s)' legal guardian/next of kin for the publication of any potentially identifiable images or data included in this article.

XY analyzed the data and wrote the paper. WT and ZZ collected and analyzed the data. YZ and CL collected the patients. XT ideated the study and revised the paper. All authors approved the final version of the manuscript.

This work was supported by the Natural Science Foundation of China (Grant Number 82001655).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.