Diagnostic efficacy of serum cytokines and chemokines in fungal bloodstream infection in febrile patients

Abstract Background The role of serum cytokines/chemokines in early diagnosis of fungal infections has not been clearly clarified yet. This study aims to measure the serum levels of cytokines/chemokines in cases of fungemia and to compare them with culture‐negative controls. Methods In total, fourteen types of serum cytokines and chemokines from 41 patients with fungemia were compared with 57 patients with negative blood culture results. The cytokine and chemokine levels were detected with multiplex platform. We then performed statistical analysis as a two‐tailed P < .05. ROC analysis was performed, and an area under the curve (AUC), and sensitivity and specificity values were calculated to determine the efficacy of various cytokines and chemokines for fungemia. Binary logistic regression was performed to further explore the combination mode of cytokines and chemokines, which could increase the diagnostic efficiency. Results C‐reactive protein and procalcitonin were significantly higher compared with those in negative control group, while white blood cell, percentage of neutrophil, percentage of lymphocyte, and ratio of neutrophil and lymphocyte did not differentiate between two groups. Serum levels of IFN‐γ, TNF‐α, MIP‐1β, IL‐6, IL‐8, IL‐10, IL‐12p70, and IL‐17 were significantly higher in patients with fungemia compared with the control group. Combination of MIP‐1β and IL‐17 could improve the AUC, sensitivity, and specificity for the diagnosis of fungemia. Conclusion Our study demonstrates that serum cytokines and chemokines including IFN‐γ, TNF‐α, MIP‐1β, IL‐6, IL‐8, IL‐10, IL‐12p70, and IL‐17 could be considered as diagnostic markers for fungemia. Combination of these biomarkers might improve the diagnostic efficiency of fungemia.


| INTRODUC TI ON
Increasing incidence of fungemia among nosocomial bloodstream infections has been observed over the recent years due to increase of immunocompromised individuals with systemic diseases such as diabetes mellitus and acquired immune deficiency syndrome. 1,2 Fungemia is a life-threatening condition with the global mortality rate surging more than fivefold in the past decade. 3,4 It has been reported that the outcome depends on early and targeted treatment; thus, prompt diagnosis is essential at the early stage of infection.
Currently, early diagnosis of fungemia remains a complicated issue. 5 Although blood culture remains the reference method for the detection of fungemia, this method takes 3-5 days to generate positive results and also suffers from low sensitivity, which could not meet the requirement of early diagnosis and clinical guidance for fungemia.
In addition, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) has been used for the diagnosis of pathogens in bloodstream infections, with a shorter turnaround time. 6 However, MALDI-TOF-MS also depends on the colony originated from blood culture, thus limiting its application. Molecular methods such as PCR assays and next-generation sequencing (NGS) have also been developed during the past years. But due to multiple manual operations in the whole process, the lack of conclusive validation for commercially available assays, and variety of methodologies, these methods have not been recommended for routine use in clinical practice. 7 Serological marker detection allows for a short turnaround time and hence is an important tool for early diagnosis of fungemia.
Cytokine analysis has been performed in bacteremia, 8 and cytokine release varies by the pattern. Studies have shown that the concentrations of cytokines/chemokines may increase earlier than currently available inflammatory proteins such as C-reactive protein (CRP) and procalcitonin (PCT). Thus, these cytokines/chemokines may be considered as diagnostic biomarkers for fungal bloodstream infection. 9 Thus, our study aims to measure the serum levels of cytokines including interleukin (IL) -1β, -2, -3, -4, -6, -8, -10, -17, -12p70, granulocyte colony-stimulating factor (G-CSF), interferon (INF)-γ, and tumor necrosis factor (TNF)-α; chemokines including macrophage inflammatory protein-1β (MIP-1β) and regulated on activation normal T cell expressed and secreted (RANTES)/chemokine (C-C motif) ligand 5 (CCL5). In addition, currently available clinical laboratory parameters including white blood cell (WBC), percentage of neutrophil (N%), percentage of lymphocyte (L%), neutrophil-to-lymphocyte ratio (NLR), C-reactive protein (CRP), and procalcitonin (PCT) were also collected from the database in cases of fungemia and to compare them with those of culture-negative individuals.

| Patient selection and group division
Patients with blood cultures that yielded fungi and patients with negative blood culture results were included in our study. All these enrolled patients were suspected of infection due to their increased temperature from fever clinics or different departments of Chinese PLA General Hospital between January 2016 and December 2018.
Informed consent was obtained from all patients. Since the number of patients with fungal bloodstream infections was not very large, these patients were consecutive. However, for the negative results patients, since the negative rate is quite high, these patients were not consecutive and we chose patients with matched age and sex with culture-positive patients. This study was reviewed and approved by the Ethics Committee of Chinese PLA General Hospital (S2018-207-02). Patients with positive blood culture results were considered as the fungemia group and patients with negative culture results as the control group.

| Blood culture and identification of pathogens
Blood samples (8-10 mL) for culture were taken from the patients for two to three times and inoculated using BACTEC PLUS Aerobic/F (BD) blood culture bottles. The isolations from blood culture were identified to species with VETIK YBC card (Bio-Merieux) and MALDI-TOF-MS (Bio-Merieux).

| Sample collection
All patients underwent blood collection during the early stage of onset (within 48 hours of admission) and before the use of antibiotics. Serum samples were obtained at the simultaneous time of blood culture and then separated immediately with the centrifuge (Thermo-Electron Corporation) at 1500 g for 5 minutes. Then, the supernatant was transferred to polypropylene tubes (Solarbio) and stored at −80° refrigerator before analysis.

| Measurement of the serum cytokine levels
The serum samples were analyzed using a Luminex ® xMAP Bead Array Platform (Millipore Corporation). A combination of 14 cytokines and chemokines was quantified using the reagents of selected ones from Human Cytokine 30-Plex Panel Kit.

| Measurement of serum CRP and PCT
CRP was measured using nephelometric method (Siemens BNII, Cardiophase), and PCT was measured by Cobas800 (Roche) based on electrochemical luminescence technology.

| Statistical analysis
Data were analyzed with SPSS 22.0 and GraphPad Prism 5 (GraphPad Software). The distribution of each continuous variable was compared with the normal distribution using the Shapiro-Wilk test. For non-normally distributed data, the median and quartiles were used to represent concentrated and discrete trends, and Mann-Whitney U test was used as a nonparametric test to compare samples between two groups. In order to determine the cutoff value for cytokine serum levels, receiver operating characteristic (ROC) analysis was performed, and an area under the curve (AUC), and sensitivity and specificity values were calculated. Statistical significance was assumed based on a value of P < .05.

| Comparison of commonly used diagnostic parameters between two groups
Commonly used diagnostic biomarkers in clinical laboratory including white blood cell (WBC), percentage of neutrophil (N%), percentage of lymphocyte (L%), and ratio of neutrophil and lymphocyte (NLR) did not differentiate between the fungemia group and the negative control group (

| Comparison of fourteen cytokines and chemokines between two groups
Fourteen cytokines and chemokines were analyzed at the early time points prior to initiation of antifungal therapy. Serum concentrations of IFN-γ, TNF-α, MIP-1β, IL-6, IL-8, IL-10, IL-12p70, and IL-17 were significantly higher in patients with the fungemia group compared with the control group, while the other six cytokines/chemokines did not differ between two groups (Table 4).  Table 5.

| D ISCUSS I ON
With the extensive use of broad-spectrum antibiotics, the increase of immunosuppressive population, and development of invasive treatment technology, the incidence of fungal bloodstream infection rises gradually in the past decade. 10  In our study, CRP and PCT significantly increased in the fungemia group compared with negative control, while other commonly used laboratory parameters such as WBC, N%, L%, and NLR did not differ between two groups. As reported by previous studies, serum CRP and PCT levels may be useful for differential diagnosis of sepsis and CRP levels were lower in patients with fungemia than those with bacteremia. 13,14 In another study, "low PCT and high CRP" were found in case of fungal infections. 15,16 In the current study, IL-17 found to be significantly increased in the fungemia group, and after combination with MIP-1β, these two cytokine/chemokines could improve the diagnostic efficiency (with the sensitivity of 81% and specificity of 81.5%). IL-17 is produced by receptor-deficient mice that showed increased susceptibility to a disseminated Candida albicans infection. 19 In addition, the importance of IL-17 for the host defense against Candida infections has been underlined by the increased number of infectious complications seen in patients with psoriasis who have been treated with IL-17A-targeted antibodies. 20 The level of cytokine IL-17 has been reported to elevate in the serum samples of patients with candidemia when compared with patients with bacterial sepsis or healthy control subjects. 21 MIP-1β is also significantly increased in our study and could be combined together with IL-17 to improve the diagnostic efficiency.
Chemokines play an important role in the trafficking of immune cells during infection. 22 The C-C chemokines include molecules include human monocyte chemotactic protein 1 (MCP-1), RANTES/CCL3, and macrophage inflammatory protein 1α and 1β (MIP-1α/CCL4 and MIP-1β/CCL5), which exhibit chemoattactant potential for monocytes but not neutrophils. 23 Through binding to its receptor (CCR5), MIP-1β regulates the balance between Treg and Th17 cells and thus  therefore, it has a pivotal role for development on proper innate immune response. 45 Romani et al reported that IL-6-deficient mice are more susceptible to disseminated candidiasis than wild-type mice, which suggests that IL-6 release is fundamental during the fungal infection. 46 In addition, Kovács R et al reported that serum IL-6 had strong relationship with systemic C albicans infection. 47 IL-8 is produced by phagocytes and mesenchymal cells exposed to inflammatory stimuli and considered as neutrophil-activating cytokine. 48 In vivo, IL-8 elicits a massive neutrophil accumulation at the site of injection during inflammation and host defense.
Zymosan induced the release of IL-6 and IL-8, which is stimulated through activating the MAPK and NF-κB pathways. 49  Tumor necrosis factor-alpha (TNF-α) is a fundamental cytokine in mounting an immune response, whereas its high levels are detrimental in inflammatory diseases. 50 TNF-α is stimulated to release after activation of phagocytosis, which plays an important role in determining the development of subsequent antifungal adaptive immune responses such as polarization of naïve Th cells into effector Th17 cells. 51,52 Studies have shown that A fumigatus could induce acute inflammation regulated by neutrophils with pro-inflammatory cytokines (TNF-α, GM-CSF, and IL-1β) and chemokines (MIP-1). 53 It has also been reported that TNF-α enhances host responses to A fumigatus, and inhibition on its function might increase susceptibility to aspergillosis. 54 In addition, anti-TNFα agents have been shown to be associated with increased infection risks for invasive fungal infections, particularly when given late in the overall course of treatment in pediatric patients. 55 The neutralization of TNF activity could lead to suppression of the production of IFN, promotion of monocyte apoptosis, and prevention of maintenance of granuloma, allowing fungus growth in several organs. 56 Previous studies have reported that IL-1β, IL-2, IL-3, IL-4, G-CSF, and RANTES play an important role in the regulating both immune activation and homeostasis. When infection occurs, these cytokines/ chemokines can lead a systemic inflammatory response. However, in this study, the levels of these cytokines/chemokines were not significantly different between two groups (P > .05).
Limitations of this study include its retrospective design and the relatively small number of available subjects. This may further limited the ability to find statistical significance among those cytokine options. In addition, the continuous variations of these cytokines were not recorded due to the absence of sample collection. Thus, we could not evaluate the correlation of cytokine changes over time.

| CON CLUS ION
In this study, we found eight serum cytokines/chemokines that significantly increased in the fungemia group when compared with the culture-negative control group. Among them, MIP-1β and IL-17 might be given more attention during the diagnosis of fungemia. The utility of these biomarkers to diagnose fungemia dynamically requires to be assessed in further studies.

ACK N OWLED G M ENTS
We would like to thank the following: Xinyu Wen helped to provide language help; Shang He helped to give suggestions concerning the culture results.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.