Non-Hodgkin’s lymphoma (NHL) is a form of tumor that originates in the lymphoid tissues. NHL patients are at significant risk of infection because of the immunosuppression induced by the malignant tumor and the extreme bone marrow suppression following tumor chemotherapy. Moreover, severe infections are mostly caused by bacteremia, which results in disrupting the therapeutic effects of the treatment and increasing the mortality rate of NHL patients (Jemal et al., 2007; Armitage et al., 2017), and is an independent prognostic criterion in lymphoma patients (Dendle et al., 2017). Infections of the lungs are common in NHL patients. Because most patients do not experience apparent symptoms during the initial stage of infection, it is difficult to detect the underlying condition before it progresses to a more critical level. To reduce the risk of death from serious infections, it is necessary to assess respiratory infections in NHL patients and develop effective early therapies. This emphasizes the importance of biomarkers in the accurate and timely diagnosis of infection and bacteremia.

Inflammatory biomarkers, including C-reactive protein (CRP) and pro-calcitonin (PCT), are routinely utilized for the diagnosis of clinical infection (Shokripour et al., 2020). CRP is an acute-phase reactant that is released in response to infection, tissue injury, and inflammation and is mediated by cytokines, such as interleukin-6, and -1 (Pepys and Baltz, 1983). Multiple factors, including malignancy, anti-inflammatory medicine, and the perioperative phase, can influence CRP levels, making it difficult to evaluate when detected outcomes fluctuate significantly (Nizri et al., 2014; Jukic et al., 2015). PCT is a pro-peptide of calcitonin, which is generally synthesized by the thyroid gland’s C cells. PCT levels in healthy people are very low (less than 0.05 ng/mL), but their levels can be considerably raised in patients with serious septic shock or sepsis (Maruna et al., 2000). Moreover, in cancer patients, the value of PCT may increase due to metastasis, the neuroendocrine function of malignant cells (Whicher et al., 2001), or the production of pro-inflammatory cytokines (MacDonald, 2007). Furthermore, imaging data can be utilized for the diagnosis of bacterial infections; although their sensitivity is limited, specifically in the initial phase of infection. Therefore, patients with malignant tumors need to be combined with other novel markers to make a more accurate diagnosis of infection.

In recent years, cytokines have gained a lot of attention as novel inflammatory indicators that perform a considerable task in the diagnosis of infection. Activated immune cells produce cytokines that enhance the host’s defense against bacterial and viral infections (Akdis et al., 2016). However, abnormal level of cytokines has been linked with the occurrence and severity of several infections. The reported studies revealed that interleukin-6 (IL-6) is a key modulator of the inflammatory response in the initial phase of sepsis, which is closely associated with organ failure (Bian et al., 2017). IL-8 and IL-6 are also abnormally raised in urinary tract infections in non-febrile patients (Rincón-López et al., 2021). Abnormal expressions of IL-6, -8, and -10 have been found in acquired pneumonia (Paats et al., 2013), and significant disruption of cytokines, such as IL-6 and -8 has also been observed in patients infected with COVID-19, which is associated with the severity of infection (Bülow Anderberg et al., 2021). The level of IL-6 can guide management of children with febrile neutropenia and are more specific and sensitive than CRP in the diagnosis of bacteremia/sepsis. IL-6 levels can differentiate gram-positive and gram-negative bacteremia and can predict the survival outcomes (Karan, 2002; Kitanovski et al., 2006; Xu et al., 2013). In addition, cytokines, including IL-6/IL-10/TNF-α/IFN-γ can be used for rapid and timely diagnosis of infection in children with hematological malignancy (Tang et al., 2011). The cytokine pattern of IL-6/IL-10 has more remarkable diagnostic precision in bacteremia and severe infection in children with hematological tumors than PCT (Xia et al., 2016). This evidence suggests that changes in inflammatory cytokines are closely related to infection. In addition, the detection methods of cytokines include real-time quantitative PCR, ELISA, and flow cytometry, among which flow cytometry is increasingly favored for its advantages of high accuracy, small sample size, and fast detection time (i.e., 3-4 hours) (Cook et al., 2001).

The therapeutic value and cost-effectiveness of various treatments should be evaluated as more technologies become available. Cytokines have previously been demonstrated to predict lung bacterial infection in recently diagnosed adult hematological malignancies (Li et al., 2021). In order to evaluate the predictive impact of cytokines for respiratory bacterial infection and respiratory infection associated with bacteremia in NHL, we examined the body temperature, cytokine profile, PCT, and CRP of 229 newly diagnosed NHL patients.

Bacterial infection is prevalent among NHL patients associated with hematological malignancies, which often accompany the whole course of the disease and even affect the survival of patients. Among them, the pulmonary bacterial infection is difficult to diagnose in the initial stage of the infection and is further complicated by a variety of factors that can quickly proceed to severe infection. About 16% of patients’ deaths are caused by pulmonary infection (Xia et al., 2016; Periselneris and Brown, 2019). Therefore, it is critical to check for infection during the initial examination. Our obtained results revealed that IL-8 and IL-6 levels were comparatively lower in the uninfected patients than in co-infected patients. Further analysis showed that IL-6 and IL-8 are the cytokines that predict respiratory infection alone and respiratory infection with bacteremia. In particular, IL-6 is a more sensitive predictor of bacterial infection compared to PCT and CRP, and different levels of IL-6 predict different extents of infection.

IL-6 expression levels are abnormally elevated during infection. In addition, T and B lymphocyte development and differentiation are dependent on the expression of IL-6. As a pro-inflammatory cytokine, IL-6 performs a critical task in initiating antibacterial inflammation and can trigger a cascade of inflammatory mediators (Bian et al., 2017). IL-6 is produced by adipocytes, fibroblasts, keratinocytes, and endothelial cells in response to leukocyte invasion, but it is also stimulated by other cytokines, including IL-1 and TNF-α. IL-6 binds to membrane-bound IL-6R, which binds to signal transduction co-receptor glycoprotein 130 (GP130) to formulate the IL-6/IL-6Rα/gp130 complex, then signal transduction occurs through the JAK-STAT pathway (Baran et al., 2018; Naseem et al., 2018). IL-6 classical signals can induce acute responses and are regarded to have anti-inflammatory and homeostatic impacts (Rose-John, 2017). Currently, IL-6 is considered to be a multifunctional cytokine, associated with various physiological processes, such as inflammation, immune response, and hematopoiesis. The steady-state IL-6 level in healthy people is about 1-10 pg/ml (Kishimoto et al., 1995), however, its serum level is significantly elevated during infection, inflammatory disease, or cancer progression and is correlated with improper disease prognosis (Taher et al., 2018; McElvaney et al., 2021). In severe sepsis, the concentration of IL-6 can reach 100–1000 ng/mL (Jones, 2005). In this study, IL-6 levels were shown to be considerably greater in non-infected NHL cases than in healthy individuals, while IL-6 levels were significantly higher in infected NHL patients and linked with infection severity. It has been revealed that IL-6 also played a significant role in differentiating various degrees of bacterial infection in NHL patients. When IL-6 ≥ 18.79 pg/ml may suggest pulmonary bacterial infection, while IL-6 ≥ 102.6pg/ml may suggest pulmonary bacterial infection with bacteremia. This is comparable with previously published results, and IL-6 has a high value in distinguishing serious infections in clinical practices (Karan, 2002; Kitanovski et al., 2006; Xu et al., 2013). It has further been shown that the disease itself does not affect the susceptibility of IL-6 to bacterial infection in NHL patients.

Patients with respiratory bacterial infections have been shown to have abnormally expressed pro-inflammatory cytokines, including IL-6, IL-8, TNF-β, and IFN-γ. Several reported studies have shown that these cytokines can be used as predictors of diverse sorts of infection. Abnormal expression of cytokines (IL-6, IL-10, IFN-γ, and TNF) are similar in children experiencing acute lower respiratory tract infection. Furthermore, these cytokines also contribute to enterovirus infection, viral encephalitis (Fuchs et al., 2018; Xu et al., 2020), and nephritis (Mizutani et al., 2017) in children. Abnormal expression of IL-6 and IL-10 has also been found in children with acquired pneumonia (Vasconcellos et al., 2018; Nascimento-Carvalho et al., 2020). TNF-α and IFN-γ are abnormally increased in bacterial infections but seem to be more significant biomarkers in Mycobacterium tuberculosis infection. The IFN-γ/TNF-α dual release fluorescence point test is a simple and effective way to diagnose active tuberculosis (Dorhoi and Kaufmann, 2014; Kim et al., 2020; Zhang et al., 2021). IL-8 mediates its signaling through extracellular binding to two G-protein-coupled receptors (GPCRs), i.e., CXC-chemokine receptor 2 (CXCR2) and CXC-chemokine receptor 1 (CXCR1). Neutrophils can be recruited through this signal and their oxidative burst and particle release can be triggered to reduce inflammatory stimuli and boost bacterial clearance rates in inflammatory sites (such as the respiratory system) (Hartl et al., 2007; Lee et al., 2020).

IL-10, an anti-inflammatory factor, is used during infection to reduce tissue damage and immunopathology through suppressing pro-inflammatory responses induced with the aid of diverse immune cells (Engelhardt and Grimbacher, 2014; Ouyang and O’Garra, 2019). In children with hematological malignancies, greater expression levels of IL-6 and IL-10 predict infection, and their high expression indicates the severity of the infection (Xu et al., 2012; Xia et al., 2016). Our previous study also proved that IL-10 is a risk factor for early death in secondary hemophagocytic syndrome cases with severe cytokine networks (Li et al., 2021). In this study, we found that IL-10 had some predictive effect on bacterial infection and respiratory bacterial infections, but its AUC did not have a strong advantage over body temperature, CRP, and PCT ( Tables 2 , 3 , and Figures 2A, B ). These evidences suggest that IL-10, IL-6, IL-8, TNF-β, and IFN-γ are all abnormally expressed in infection. Particularly, IL-6, IL-10, and IL-8 can serve as predictive markers of infection. In brief, the development of an inflammatory response involves the activation of several cytokines (Karan, 2002; Kitanovski et al., 2006; Tang et al., 2011; Xu et al., 2013). Monitoring the variations in cytokine levels of NHL patients is critical for guiding the prevention and treatment of the underlined infection.

We emphasized the importance of IL-6 and IL-8 in bacterial infections. Simultaneously, we suggest that patients’ cytokine levels, notably IL-6, be assessed at the time of diagnosis to help detect bacterial infection in NHL patients. Our research has following limitations: 1. Our data is a single-center study; 2. We only included patients with bacterial infections and excluded data on fungal and viral infections since fungal and viral infections were discovered in fewer cases at the time of their initial diagnosis. This will be our focus in the future.

Our findings imply that an aberrant rise in many cytokines in newly diagnosed NHL cases can predict the presence of various levels of bacterial infection. IL-6, in particular, has superior sensitivity and accuracy to body temperature, PCT, and CRP. In newly diagnosed NHL patients, IL-6 ≥18.79pg/mL could indicate the existence of pulmonary bacterial infection while IL-6 ≥102.6pg/mL may suggest pulmonary bacterial infection with bacteremia. Although IL-6 level in co-infected NHL patients was significantly higher than non-infection patients and healthy people. And the benefits of IL-6 in detecting infection in NHL cases were also verified. It’s also been proposed that NHL itself may not affect the susceptibility of sensitivity to bacterial infection. Briefly, these findings may aid in the identification and distinction of lung bacterial infection and the presence of bacteremia in newly diagnosed NHL patients, as well as suggest antibiotic treatment.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by The Ethics Committee of The First People’s Hospital of Yunnan Province. The patients/partici5pants provided their written informed consent to participate in this study.

All authors participated in the research and agreed to the publishing license. LZ and JPZ analyzed the data and drafted the manuscript. ZL, RZ, and XG designed the research plan, reviewed the manuscript and revised the main content. XL, FL, ZY, SZ, JLZ, and HL statistics. TY, KS, and HH checked the logic of the manuscript. All authors contributed to the article and approved the submitted version.

The author would like to thank the patients who participated in the treatment at the First People’s Hospital and Kunming Junmai Technology Co. Ltd. for providing reagents and consumables. This work was supported by the National Natural Science Foundation of China (82060810) and Open Project of Yunnan Blood Clinical Medical Center (2019LCZXKF-XY11, 2019LCZXKF-XY03, 2019LCZXKF-XY09, 2020LCZXKF-XY07 and 2021LCZXXF-XY09).

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.