Cytokine profiles as novel diagnostic markers of Epstein-Barr virus–associated hemophagocytic lymphohistiocytosis in children

    Cytokine profiles as novel diagnostic markers of Epstein–Barr virus-associated haemophagocytic lymphohistiocytosis in children Xiu-cui Han, Qing Ye, Wei-ying Zhang, Yong-min Tang, Xiao-jun Xu, Ting Zhang PII: DOI: Reference:

S0883-9441(16)31012-7 doi: 10.1016/j.jcrc.2017.02.018 YJCRC 52418

To appear in:

Journal of Critical Care

Please cite this article as: Han Xiu-cui, Ye Qing, Zhang Wei-ying, Tang Yong-min, Xu Xiao-jun, Zhang Ting, Cytokine profiles as novel diagnostic markers of Epstein–Barr virus-associated haemophagocytic lymphohistiocytosis in children, Journal of Critical Care (2017), doi: 10.1016/j.jcrc.2017.02.018

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Cytokine profiles as novel diagnostic markers of Epstein-Barr virus-associated haemophagocytic lymphohistiocytosis in children *

1

RI P

T

Xiu-cui Han1·Qing Ye1 ·Wei-ying Zhang2·Yong-min Tang1·Xiao-jun Xu1 ·Ting Zhang3 Zhejiang Key Laboratory for Neonatal Diseases, Clinical Laboratory, The Children’s Hospital of

SC

Zhejiang University School of Medicine, Hangzhou, PR China 2

Clinical Laboratory, Hangzhou first people’s Hospital, Hangzhou, PR China

3



MA NU

Medical Technology, Zhejiang Chinese Medical University, Hangzhou, PR China

Corresponding author: Qing Ye. Children’s Hospital of Zhejiang University School of Medicine, #

ED

3333 Binsheng Road, Hangzhou 310051, PR China. Tel: +86 571 87061007; Fax: +86 571 87033296;

CE

Abstract

PT

E-mail: [email protected].

AC

Purpose：The aim of this study was to identify specific laboratory indices to distinguish Epstein-Barr virus-associated haemophagocytic lymphohistiocytosis (EBV-HLH) in children. Materials and Methods: In this prospective study, Th1/Th2 cytokines, including IL-2, IL-4, IL-6, IL-10, TNF-α and IFN-γ, were analysed in patients with EBV-HLH or sepsis at the onset of disease by flow cytometry. Results: IL-10, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 were higher and IL-6 was lower in EBV-HLH patients compared to sepsis patient levels. When using the criteria of IL-10 >20.9 pg/mL, IFN-γ >17.9 pg/mL, IL-10/IL-6 >0.5 and IFN-γ/IL-6 >0.7, the sensitivity was 89.8%, 93.2%, 93.2% and 91.5%, while the specificity was 89.8%, 100%, 94.9% and 100%, respectively. After treatment of EBV-HLH patients, IL-6, IL-10, TNF-α and IFN-γ were significantly reduced (IL-6: P < 0.001; IL-10: P < 0.001;

1

ACCEPTED MANUSCRIPT TNF-α: P = 0.011; IFN-γ: P < 0.001). Conclusions: This study showed that IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 are novel specific indicators for differential diagnosis of EBV-HLH. Additionally, IL-6,

RI P

T

IL-10, TNF-α, and IFN-γ are useful indices for monitoring the effects of treatment on EBV-HLH. Keywords EBV-HLH· cytokines· IL-10/IL-6· IFN-γ/IL-6

SC

Introduction

Haemophagocytic lymphohistiocytosis (HLH) is a hyperinflammatory syndrome characterized by

MA NU

fever, organomegaly, and cytopaenia [1] and is also indicative of macrophage activation [2,3]. Aberrant activation of T cells, natural killer (NK) cells and macrophages results in hypercytokinaemia (i.e., a cytokine storm), which can cause multiorgan failure [4]. HLH is generally divided into two types:

ED

primary HLH and secondary HLH. The former is caused by genetic defects, such as PRF1 (which

PT

encodes perforin) [1,5]. Infections, malignancies, or rheumatological diseases are triggers of secondary HLH [6]. If treatment is not timely, the mortality rate of HLH in children is very high [4,7].

CE

Epstein–Barr virus (EBV) is present worldwide [8]. The occurrence of HLH in the absence of a

AC

genetic defect is frequently related to EBV, particularly in far east Asia [9]. EBV has been identified as the major triggering factor in infection-associated HLH cases [10]. Prompt and accurate treatment can result in a good outcome, but due to the nonspecific clinical manifestations and laboratory findings, clinicians have difficulty distinguishing EBV-HLH from other diseases, such as sepsis [4,11]. For improvement of survival and prognosis, early diagnosis is urgently needed. Thus, we enrolled EBV-positive patients with HLH into this study to identify new diagnostic criteria. Methods Patients

2

ACCEPTED MANUSCRIPT This prospective study was conducted from January 2013 to December 2014 in the Children’s Hospital of Zhejiang University School of Medicine. It was approved by the medical ethics committee

RI P

T

of the Children’s Hospital of Zhejiang University School of Medicine. Informed written consent was obtained from guardians on behalf of the participants involved in the study. The diagnostic criteria of

SC

EBV-HLH were as follows: the patient has a HLH diagnosis and is EBV-positive. The HLH diagnosis was made according to the HLH-2004 protocol [12]. Briefly, EBV-HLH patients were positive for the

MA NU

EBV genome in peripheral blood, bone marrow, or other tissues and had clinical symptoms compatible with HLH, such as persistent high fever, hepatosplenomegaly, cytopaenias, hypertriglyceridaemia, coagulopathy with hypofibrinogenaemia, liver dysfunction, elevated ferritin levels, serum

ED

transaminases, and neurological symptoms. Gene sequencing analysis of all the EBV-HLH patients

PT

was performed to exclude primary HLH, and a gene panel (PRF1, UNC13D, STX11, STXBP2, SH2D1A, BIRC4, RAB27A, AP3B1 and LYST) was used. The treatment was based on the HLH-2004

CE

treatment protocol described in the literature [12]. Briefly, the initial therapy consisted of

AC

dexamethasone (DXM, 10 mg/m2 for 2 weeks followed by 5 mg/m2 for 2 weeks, 2.5 mg/m2 for 2 weeks, 1.25 mg/m2 for 1 week, and 1 week of tapering), etoposide (150 mg/m 2 twice weekly for the first 2 weeks and then weekly), and cyclosporine A (6 mg/kg/d orally, aiming at blood level of 200 µg/l). After 8 weeks of the initial treatment, patients underwent continuous therapy, starting from week 9, which consisted of DXM pulses of 10 mg/m2 for 3 d and etoposide infusions of 150 mg/m2 every alternative week for a total of 12 months. The sepsis criterion used was that established by the international paediatric sepsis consensus [13]. Bacterial bloodstream infections were documented by microbiological culture at the onset of sepsis. They were considered pathogens when one of the following criteria was met: (a) two specimens from different sites had the same results; (b) if the other

3

ACCEPTED MANUSCRIPT sample was negative, a PCR assay of the blood sample was positive for gram-positive bacteria; (c) the same results were obtained from a repeated culture. All blood samples were taken for analysing serum

RI P

T

Th1/Th2 cytokines when patients were enrolled in the study. Cytokine determination

SC

Blood samples were centrifuged at 1,000 ×g at 20°C for 20 min after clotting. The serum was carefully harvested, and Th1/Th2 cytokine levels were measured immediately using 320 flow

MA NU

cytometry. The Th1/Th2 cytokines, including IL-2, IL-4, IL-6, IL-10, TNF-α and IFN-γ, were measured with a CBA kit (CBA Human Th1/Th2 Cytokine Kit II; BD Biosciences, San Jose, California) [14,15].

ED

Statistical analysis

PT

The Mann-Whitney U test and Kruskal-Wallis test were used to compare the differences in continuous variables. The χ2 or Fisher’s exact test was used to compare the differences in categorical

CE

variables. All statistical analyses were performed using SPSS version 18.0. P < 0.05 was considered to

AC

be statistically significant. The receiver operating characteristic (ROC) curve was used to assess the diagnostic value of IL-2, IL-4, IL-6, IL-10, TNF-α, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 to discriminate EBV-HLH from sepsis by MedCalc 9.4.2.0 software. The optimal diagnosis threshold was determined according to the Youden index J, and the relative sensitivity, specificity and positive and negative predictive values were calculated. Results Characteristics of EBV-HLH patients Of the 118 EBV-HLH patients, 60 were boys and 58 were girls. All of these patients had a fever and hyperferritinaemia (> 500 µg/L). Apparent peripheral blood cytopaenia (neutropaenia < 1.0ｘ109/L:

4

ACCEPTED MANUSCRIPT 83.0%; anaemia < 90 g/L: 82.0%; thrombocytopenia < 100 ｘ 109/L: 92.0%), liver dysfunction (hypertriglyceridaemia≥3.0 mmol/L: 82.0%; hypofibrinogenaemia ≤1.5 g/L: 83.0%), splenomegaly

RI P

T

(85.0%) and haemophagocytosis (91.0%) were observed in most of the patients. Additionally, 85.0% of these patients had impaired NK cell activity. Of the 120 sepsis patients, 61 were boys and 59 were girls.

SC

All of these patients had a fever. Splenomegaly (86.7%), impaired NK cell activity (78.3%), hyperferritinaemia >500 µg/L (75.8%) and anaemia < 90 g/L (62.5%) were observed in most of the

MA NU

patients. Neutropaenia < 1.0 ｘ 109/L (5.0%), thrombocytopenia of 100 ｘ 109/L (37.5%), hypertriglyceridaemia ≥ 3.0 mmol/L (6.7%), hypofibrinogenaemia ≤ 1.5 g/L (12.5%) and haemophagocytosis (0.0%) were uncommon among sepsis patients. The details are shown in Table 1.

ED

Cytokine profiles of the patients

PT

(1) Cytokine levels in patients with sepsis

The median levels and range of different cytokines in the sepsis group are shown in Table 2 and Fig.

CE

1. Compared with the normal control group, the sepsis group had elevated levels of IL-4, IL-6, IL-10,

AC

IFN-γ and TNF-α to different extents, with IL-6 significantly increased (median levels, pg/mL: IL-4: 3.5 vs. 2.8; IL-10: 10.4 vs. 3.6; IFN-γ: 8.1 vs. 6.7; TNF-α: 3.2 vs. 3.0; IL-6: 190.7 vs. 3.8). However, the expression of IL-2 was similar (median level, pg/mL: IL-2: 4.0 vs. 4.2). (2) Cytokine levels in patients with EBV-HLH The median levels and ranges of the cytokines in the EBV-HLH group before clinical treatment are shown in Table 2 and Fig. 1. In the patients with EBV-HLH, IL-4 and TNF-α were slightly elevated, with concentrations of 3.3 pg/ml and 3.3 pg/ml, respectively, compared with the normal control concentrations of 2.8 pg/ml and 3.0 pg/ml, respectively. IL-6, IL-10, and IFN-γ were significantly increased in patients with EBV-HLH compared to those of the normal control (median levels, pg/mL:

5

ACCEPTED MANUSCRIPT IL-6: 35.3 vs. 3.8, P<0.001; IL-10: 295.2 vs. 3. 6, P < 0.001; IFN-γ: 656.6 vs. 6.7, P < 0.001). The

RI P

(3) Differential expression of cytokines between EBV-HLH and sepsis

T

expression of IL-2 was similar in EBV-HLH.

The expression of the 6 cytokines differed between EBV-HLH and sepsis patients. IL-6 was

SC

moderately elevated, with a median concentration of 35.3 pg/mL in EBV-HLH, which was lower than that in patients with sepsis (median levels, pg/mL: IL-6: 35.3 vs. 190.7, p<0.05). The concentrations of

MA NU

IL-10 and IFN-γ (295.2 pg/mL and 656.6 pg/mL) were much higher than those in patients with sepsis (median levels, pg/mL: IL-10: 295.2 vs. 10.4, P < 0.001; IFN-γ: 656.6 vs. 8.1, P < 0.001). Most interestingly, the study found that in EBV-HLH, IFN-γ/IL-6 and IL-10/IL-6, with median

ED

concentrations of 13.8 and 5.4, respectively, had much higher values than those in patients with sepsis

PT

(median levels: IFN-γ/IL-6: 13.8 vs. 0.1, P < 0 .001; IL-10/IL-6: 5.4 vs. 0.1, P < 0 .001). The details are shown in Fig. 1 and Table 2.

CE

Diagnostic accuracy of the cytokines in EBV-HLH

AC

As shown above, the cytokines were differentially expressed in EBV-HLH and sepsis patients. We evaluated the abilities of these indicators to discriminate EBV-HLH from sepsis by ROC analysis (Fig. 3). The areas under the ROC curve (AUC) for IL-10, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 were 0.957, 0.990, 0.976 and 0.985, respectively, indicating that IL-10, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 were significant indicators for patients with EBV-HLH. When cut-off values for IL-10, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 were set as > 20.9 pg/mL, > 17.9 pg/mL, > 0.5 and > 0.7, the sensitivity was 89.8%, 93.2%, 93.2% and 91.5%, and the specificity was 89.8%, 100.0%, 94.9% and 100.0%, respectively. The details are shown in Fig. 3 and Table 2. Cytokine levels in patients with EBV-HLH after treatment

6

ACCEPTED MANUSCRIPT After clinical treatment, the reduction in IL-2 and IL-4 was not statistically significant (IL-2: P = 0.367; IL-4: P =0.375). IL-6, IL-10, IFN-γ and TNF-α were decreased significantly (IL-6: P < 0.001;

RI P

T

IL-10: P < 0.001; IFN-γ: P < 0.001; TNF-α: P = 0.011). The details are shown in Fig. 2. Discussion

SC

Pathogenesis of EBV-HLH is related to dysfunction of Th1 and Th2 cells, which are CD4+ T cells. Th1 and Th2 cells have different cytokine profiles. Th1 cells secrete IL-2, IL-12, IFN-γand TNF-α,

MA NU

while IL-4, IL-5, IL-6, IL-9, IL-10 and IL-13 are secreted by Th2 cells. The cytokines have effector functions and are important regulators of T cells [16]. The dysregulation of Th1 and Th2 cells results in a cytokine storm [17], which leads to most of the observed clinical manifestations and laboratory

ED

changes [1]. In this study, infiltration of lymphocytes and histiocytes led to splenomegaly (85%) and

PT

hepatomegaly, which resulted in hypertriglyceridaemia (82%), hypofibrinogenaemia (83%) and hyperferritinaemia (100%), as observed in an earlier study [18]. Additionally, all patients with

CE

EBV-HLH had a fever caused primarily by high levels of interleukins. In our study, IL-4 showed a

AC

slight increase (median levels, pg/mL: 3.3 vs. 2.8), while IL-2 and TNF-α were barely changed in EBV-HLH. These findings are similar to those of a previous study [19]. IL-6 was increased more in sepsis than that in EBV-HLH, while IL-10 and IFN-γ were significantly higher in EBV-HLH than that in sepsis (median levels, pg/mL: IL-6: P < 0.05; IL-10: P < 0.001; IFN-γ: P < 0.001). IL-6, IL-10 and IFN-γ had relatively high sensitivity and specificity (sensitivity: 49.2%, 89.8% and 93.2%; specificity: 88.1%, 89.8% and 100.0%, respectively). Among them, IL-6 can induce fever [20] and amplify acute inflammation [21,22], and IL-6 combined with a soluble form of IL-6R (sIL-6R) forms a conjugate that prolongs the half-life of IL-6. Additionally, IL-6 contributed to the transition into the chronic phase of inflammation [23]. IL-10 is a pleiotropic cytokine with anti-inflammatory activities. IL-10 inhibits the

7

ACCEPTED MANUSCRIPT release of pro-inflammatory cytokines on antigen-presenting cells [24,25]. It can directly inhibit cytokine production and T-cell function [26] and proliferation [27] to limit inflammation. IFN-γ

RI P

T

contributes to liver impairment, while IL-10 is related to cytopaenias [28]. The existing studies show that T-cell suppression is an important factor in infection. Although EBV-HLH was characterized by

SC

hyperinflammation and involved a viral infection, it was different from bacterial infections such as sepsis. Thus, IL-6, IL-10 and IFN-γ are the useful indicators for ruling out sepsis. High IL-6 levels with

MA NU

a low IFN-γ concentration would suggest a bacterial infection [29].

Because of the differing levels of IL-6, IL-10 and IFN-γ, we established ratios of IFN-γ/IL-6 and IL-10/IL-6 to observe the difference between EBV-HLH and sepsis. We found that IFN-γ/IL-6 and

ED

IL-10/IL-6 were notably different between EBV-HLH and sepsis. From the ROC curve, we observed

PT

that the two indexes had high sensitivity and specificity (sensitivity: IFN-γ/IL-6: 91.5%; IL-10/IL-6:

from infection.

CE

93.2%; specificity: IFN-γ/IL-6: 100.0%; IL-10/IL-6: 94.9%) in the differential diagnosis of EBV-HLH

AC

We also determined the cytokine profiles of the EBV-HLH patients after clinical treatment and found that IL-2, IL-4, IL-6, IL-10, TNF-α and IFN-γ were reduced to different degrees. The decrease in IL-6, IL-10, TNF-α and IFN-γ was statistically significant (IL-6: P < 0.001; IL-10: P < 0.001; IFN-γ: P < 0.001; TNF-α: P = 0.011). Thus, we hypothesized that IL-6, IL-10, TNF-α and IFN-γ are useful indices for monitoring treatment of EBV-HLH. This study has some limitations. The most important limitation is that sCD25 testing was not available, and levels were not measured. sCD25 as a activator of T-lymphocytes is an important and sensitive marker in HLH. Because the activated T-lymphocytes can secrete large amounts of IFN-γ, and

8

ACCEPTED MANUSCRIPT another study showed the correlation of sCD25 with IFN-γ [30], we considered that the IFN-γ level reflected sCD25 count.

RI P

T

In conclusion, hypercytokinaemia is quite common in various kinds of serious infections, such as sepsis caused by gram-negative bacteria, in clinical practice. However, the findings in this paper

SC

revealed that the cytokine expression profile in HLH was completely different from that in sepsis although they both show hypercytokinaemia. IFN-γ and IL-10 were substantially elevated, while IL-6

MA NU

was slightly elevated in HLH; however, in sepsis, IL-6 was substantially elevated, IL-10 was even higher, while IFN-γ was normal or slightly elevated, and the elevation of TNF-α of some patients was significant. In addition, IFN-γ was increased slightly in viral infections, mostly below 100 ng/L

ED

(according to the analysis in this research). In systemic inflammatory response syndrome, IL-6 and

PT

IL-10 are increased, while TNF-α and IFN-γ were not elevated. Thus, this research indicates that the variations in IFN-γ, IL-10 and IL-6 are highly specific for HLH diagnosis and can be used to identify

CE

HLH and other diseases, especially infectious diseases.

AC

Acknowledgement We thank the children and parents who participated in this study for their time and generosity. This project was supported by grants from the National Natural Science Foundation of China (Grant No. 81501760), Zhejiang Provincial Natural Science Foundation of China (Grant No. LQ16H050002), and Zhejiang Provincial Healthy Science Foundation of China (Grant No. 2015KYB191). The funders did not take part in the study. Compliance with ethical standards Conflict of interests The authors declare no conflict of interests.

References

9

ACCEPTED MANUSCRIPT 1. Janka GE, Lehmberg K (2014) Hemophagocytic syndromes--an update. Blood Rev 28 (4):135-142. doi:10.1016/j.blre.2014.03.002 2. Lemerle S, Bernaudin F, Papay-Paillerets I, Lobut JB, Doppelt E, Reinert P (1989) Macrophage

RI P

T

activation syndrome linked to Epstein-Barr virus. Ann Pediatr (Paris) 36 (8):539-543 3. Stephan JL, Zeller J, Hubert P, Herbelin C, Dayer JM, Prieur AM (1993) Macrophage activation syndrome and rheumatic disease in childhood: a report of four new cases. Clin Exp Rheumatol 11

SC

(4):451-456

4. Kaya Z, Bay A, Albayrak M, Kocak U, Yenicesu I, Gursel T (2015) Prognostic Factors and

MA NU

Long-Term Outcome in 52 Turkish Children With Hemophagocytic Lymphohistiocytosis. Pediatr Crit Care Med 16(6):e165-173

5. Stepp SE, Dufourcq-Lagelouse R, Le Deist F, Bhawan S, Certain S, Mathew PA, Henter JI, Bennett M, Fischer A, de Saint Basile G, Kumar V (1999) Perforin gene defects in familial

ED

hemophagocytic lymphohistiocytosis. Science 286 (5446):1957-1959 6. Tothova Z, Berliner N (2015) Hemophagocytic Syndrome and Critical Illness: New Insights Diagnosis

and

Management.

J

Intensive

Care

Med

30

(7):401-412.

PT

into

doi:10.1177/0885066613517076

CE

7. Bin Q, Gao JH, Luo JM (2016) Prognostic factors of early outcome in pediatric hemophagocytic lymphohistiocytosis: an analysis of 116 cases. Ann Hematol 95(9):1411-1418.

AC

doi: 10.1007/s00277-016-2727-6 8. Hislop AD, Taylor GS, Sauce D, Rickinson AB (2007) Cellular responses to viral infection in humans:

lessons

from

Epstein-Barr

virus.

Annu

Rev

Immunol

25:587-617.

doi:10.1146/annurev.immunol.25.022106.141553 9. Imashuku S (2002) Clinical features and treatment strategies of Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. Crit Rev Oncol Hematol 44 (3):259-272 10. IshiiE, OhgaS, ImashukuS (2007) Nationwide survey of hemophagocytic lymphohistiocytosis in Japan[J]. Int J Hematol 86(1)：58–65 11. Du L, Pan WH, Cai W, Wang J, Wu YM, Shi CR (2014) Delivery room surgery: an applicable therapeutic strategy for gastroschisis in developing countries. World Journal of Pediatrics 10 (1):69-73. doi:10.1007/s12519-014-0455-3 10

ACCEPTED MANUSCRIPT 12. Henter JI, Horne A, Arico M, Egeler RM, Filipovich AH, Imashuku S, Ladisch S, McClain K, Webb D, Winiarski J, Janka G (2007) HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic

lymphohistiocytosis.

Pediatr

Blood

Cancer

(2):124-131.

RI P

T

doi:10.1002/pbc.21039

48

13. Goldstein B, Giroir B, Randolph A (2005) International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med 6 (1):2-8.

SC

doi:10.1097/01.pcc.0000149131.72248.e6

14. Ye Q, Shao WX, Shang SQ, Shen HQ, Chen XJ, Tang YM, Yu YL, Mao JH (2016) Clinical

MA NU

Value of Assessing Cytokine Levels for the Differential Diagnosis of Bacterial Meningitis in a Pediatric Population. Medicine (Baltimore) 95 (13):e3222. doi:10.1097/md.0000000000003222 15. Ye Q, Shao WX, Xu XJ, Yang YZ (2014) The clinical application value of cytokines in treating infectious diseases. PLoS One 9 (6):e98745. doi:10.1371/journal.pone.0098745

ED

16. Sharma R, Fu SM, Ju ST (2011) IL-2: a two-faced master regulator of autoimmunity. J Autoimmun 36 (2):91-97

PT

17. Szyper-Kravitz M (2009) The hemophagocytic syndrome/macrophage activation syndrome: a final common pathway of a cytokine storm. Isr Med Assoc J 11 (10):633-634

CE

18. Janka GE (2007) Familial and acquired hemophagocytic lymphohistiocytosis. Eur J Pediatr 166 (2):95-109. doi:10.1007/s00431-006-0258-1

AC

19. Tang Y, Xu X, Song H, Yang S, Shi S, Wei J, Pan B, Zhao F, Liao C, Luo C (2008) Early diagnostic and prognostic significance of a specific Th1/Th2 cytokine pattern in children with haemophagocytic syndrome. Br J Haematol 143 (1):84-91 20. Veldhuis GJ, Willemse PH, Sleijfer DT, van der Graaf WT, Groen HJ, Limburg PC, Mulder NH, de Vries EG (1995) Toxicity and efficacy of escalating dosages of recombinant human interleukin-6 after chemotherapy in patients with breast cancer or non-small-cell lung cancer. J Clin Oncol 13 (10):2585-2593 21. Hirano T (1992) Interleukin-6 and its relation to inflammation and disease. Clin Immunol Immunopathol 62 (1 Pt 2):S60-65 22. Fonseca JE, Santos MJ, Canhao H, Choy E (2009) Interleukin-6 as a key player in systemic inflammation

and

joint

destruction.

doi:10.1016/j.autrev.2009.01.012 11

Autoimmun

Rev

8

(7):538-542.

ACCEPTED MANUSCRIPT 23. Atreya R, Mudter J, Finotto S, Mullberg J, Jostock T, Wirtz S, Schutz M, Bartsch B, Holtmann M, Becker C, Strand D, Czaja J, Schlaak JF, Lehr HA, Autschbach F, Schurmann G, Nishimoto N, Yoshizaki K, Ito H, Kishimoto T, Galle PR, Rose-John S, Neurath MF (2000) Blockade of

RI P

T

interleukin 6 trans signaling suppresses T-cell resistance against apoptosis in chronic intestinal inflammation: evidence in crohn disease and experimental colitis in vivo. Nat Med 6 (5):583-588. doi:10.1038/75068

SC

24. de Waal Malefyt R, Abrams J, Bennett B, Figdor CG, de Vries JE (1991) Interleukin 10(IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by

MA NU

monocytes. J Exp Med 174 (5):1209-1220

25. Fiorentino DF, Zlotnik A, Mosmann TR, Howard M, O'Garra A (1991) IL-10 inhibits cytokine production by activated macrophages. J Immunol 147 (11):3815-3822 26. de Waal Malefyt R, Yssel H, de Vries JE (1993) Direct effects of IL-10 on subsets of human

Immunol 150 (11):4754-4765

ED

CD4+ T cell clones and resting T cells. Specific inhibition of IL-2 production and proliferation. J

PT

27. Groux H, Bigler M, de Vries JE, Roncarolo MG (1996) Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells. J Exp Med 184 (1):19-29

CE

28. Yang SL, Xu XJ, Tang YM, Song H, Xu WQ, Zhao FY, Shen DY (2016) Associations between inflammatory cytokines and organ damage in pediatric patients with hemophagocytic

AC

lymphohistiocytosis. Cytokine 85:14-17. doi:10.1016/j.cyto.2016.05.022 29. Xu XJ, Tang YM, Song H, Yang SL, Xu WQ, Zhao N, Shi SW, Shen HP, Mao JQ, Zhang LY, Pan BH (2012) Diagnostic accuracy of a specific cytokine pattern in hemophagocytic lymphohistiocytosis in children. J Pediatr 160 (6):984-990 e981. doi:10.1016/j.jpeds.2011.11.046 30. Imashuku S, Hibi S, Sako M, Ishida Y, Mugishima H, Chen J, et al. (1995) Soluble interleukin-2 receptor: a useful prognostic factor for patients with hemophagocytic lymphohistiocytosis. Blood 86:4706-4707.

12

ACCEPTED MANUSCRIPT Figure legends Fig. 1 Comparisons of serum cytokine concentrations among patients with EBV-HLH or sepsis and

RI P

T

normal controls.

The median levels of IL-2 in the three groups were 3.7, 4.0, and 4.2 pg/mL, respectively; IL-4, 3.3,

SC

3.5,and 2.8 pg/mL; IL-6, 35.3, 190.7, and 3.8 pg/mL; IL-10, 295.2, 10.4, and 3.6 pg/mL; TNF-α, 3.3, 3.2, and 3.0 pg/mL; IFN-γ, 656.6, 8.1, and 6.7 pg/mL; IFN-γ/IL-6, 13.8, 0.1, and 2.2; IL-10/IL-6, 5.4,

MA NU

0.1, and 1.1, respectively.

Note: “HLH” in the figure is consistent with the “EBV-HLH” in the text.

ED

Fig. 2 Comparisons of serum cytokine concentrations (pg/mL) in patients with EBV-HLH before and

PT

after clinical treatment.

CE

The cytokines are IL-2, IL-4, IL-6, IL-10, TNF-α and IFN-γ.

AC

Fig. 3 ROC curve for the predictive value of IL-10, IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 in patients with EBV-HLH or sepsis.

13

ACCEPTED MANUSCRIPT

T

Table 1. The Features of EBV-HLH and Sepsis EBV-HLH

RI P

Features Demography Number Gender（boys/girls） clinic manifestation Fever Splenomegaly

SC

118 60/58

MA NU

Neutropenia < 1.0 × 109/L Anemia < 90 g/L

AC

CE

PT

ED

Thrombocytopenia<100×109／L Hypertriglyceridemia ≥3.0 mmol/L， Hypofibrinogenemia ≤1.5 g/L Hyperferritinemia >500µg/L Hemophagocytosis Impaired NK cell activity

14

sepsis 120 61/59

100.0% 85.0%

100.0% 86.7%

83.0% 82.0%

5.0% 62.5%

92.0% 82.0% 83.0% 100.0% 91.0% 85.0%

37.5% 6.7% 12.5% 75.8% 0.0% 78.3%

ACCEPTED MANUSCRIPT

Table 2. Performances of Inflammatory Cytokines in Discriminating EBV-HLH HLH

Sensiti AU sepsis

control

Youden in

Cuto

P

Indicators

PP Specificit

vity C

dex J

ff

y (%)

（1.1-7.4

4.2

0.0

(3.0-7.0）

(2.0-8.0）

07

3.3

3.5

(2.0-11.1

2.8

（1.9-6.2）

0.0

（ 2.0-6.

00

0）

IL-6（pg/mL）

(3.6-207

190.7

3.8

（ 9.7-504

0.0

（ 1.1-42

0.373

0.6

5.0）

.4）

295.2

10.4

3.6

IL-10（pg/mL） (11.3-500

（ 3.3-213.

（ 2.6-12

0.0

9)

.0)

3.3

3.2

3.0

（pg/mL）

(1.5-13.2

（2.0-5.0）

)

（ 2.4-305.

71.9 1

≤

80. 49.2

88.1

34.0

63.4 6

>20.

00

0.0

89. 89.8

89.8

9

89.8 8

0.282

0.6

04

77. >3.5

40.7

88.1

02

6.7

0.0

（ 1.5-10

00

8)

.2)

5.4

0.1

1.1

CE

00.0)

(0.1-180.

AC

13.8 IFN-γ/ IL-6

8.1

PT

656.6

5)

39.0

59.8 4

2）

IFN-γ（pg/mL） (12.2-50

IL-10/ IL-6

（ 1.0-4.

ED

TNF-a

58. 84.7

0.797

0.9 57

0.0)

59.6 .0

≤

00

97

8.5)

100

4.3

MA NU

35.3

(%) (%)

100.0

3.0

0.237 0.5 86

)

≤

32.2

79

）

IL-4（pg/mL）

0.322 0.6

SC

IL-2（pg/mL）

4.0

RI P

(%) 3.7

NPV V

T

Diagnostic

(0.1-279.

(0.01-0.83)

(0.4-1.7)

0.932 0.9

>17.

100 93.2

90 0.0

100.0

9

93.7 .0

0.881 0.9

94.

00

>0.5

93.2

94.9

76

0.1

2.2

0.0

0.9

(0.01-0.8)

(0.3-3.8)

00

85

93.3 8

0.915 100 >0.7

91.5

100.0

92.2 .0

3)

PPV = positive predictive value, NPV = negative predictive value, AUC = area under the curve. The p-values refer to the differences of EBV-HLH, sepsis and healthy control.

15

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Figure 1

16

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Figure 2

17

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

Figure 3

18

ACCEPTED MANUSCRIPT IFN-γ, IL-10/IL-6 and IFN-γ/IL-6 are new, specific indicators for the differential diagnosis of EBV-HLH. IL-6, IL-10, TNF-α, and IFN-γ are useful indices for the observation of a curative

AC

CE

PT

ED

MA NU

SC

RI P

T

effect in EBV-HLH.

19