Myeloproliferative neoplasms (MPN) are hematological disorders characterized by increased proliferation and accumulation of mature myeloid cells in the bone marrow and peripheral blood. MPN are classified as Philadelphia (Ph) chromosome-positive or -negative. The present investigation focusses on three Ph-negative MPN: essential thrombocythemia (ET), primary myelofibrosis (PMF), and polycythemia vera (PV).1,2 These MPN are clonal disorders that result from malignant transformations of hematopoietic stem cells, which raise the number of mature myeloid blood cells, especially megakaryocytes/platelets in ET, erythrocytes in PV, and both megakaryocytes/platelets and granulocytes in PMF. Nearly 10–15% of ET, PMF, and PV patients progress to acute myeloid leukemia.1,2

Constitutive activation of the Janus kinase 2 (JAK2)/signal transducer and activator of transcription (STAT) signaling pathway due to acquired somatic mutations in the JAK2, calreticulin or MPL genes may drive the course of MPN. The JAK2 V617F mutation, which is detected in more than 90% of PV patients and in about half of ET and PMF patients, is located in the pseudokinase domain of the JAK2 protein and elicits a ‘pre-activation state’.3,4 This mutation seems to confer the advantage of survival and proliferation to myeloid hematopoietic cells independently of stimulatory signals, leading to clonal expansion of myeloid progenitors and mature cells.4 In MPN, the hematopoietic cells display an exacerbated response to cytokines and growth factors such as erythropoietin, granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin 3 (IL-3), and insulin-like growth factor-1 (IGF-1).5 Autocrine stimulation may occur due to either defective cytokine receptors or constitutive activation of the JAK2, STAT5, phosphoinositide 3-kinase (PI3K), signal-regulated kinases extracellular (ERK), and protein kinase B (PKB) signaling pathways.5

MPN are considered onco-inflammatory disorders. The concept of ‘onco-inflammation’ proposes a crosstalk between cancer cells and their inflammatory microenvironment suggesting that this plays a relevant role in the initiation and progression of cancers.6,7 MPN are characterized by abnormal immune system activity, increased monocyte/macrophage compartment, expansion of myeloid-derived suppressor cells, abnormal frequency of regulatory T cells, and dysfunction of natural killer and T CD4+ cells.8 In line with this concept, chronic inflammation in MPN has been described as a potential facilitator for premature atherosclerosis development and clonal evolution to acute myeloid leukemia.9 In addition, fibrotic transformation, commonly found in PMF patients and in myelofibrosis secondary to ET and PV, seems to be associated with an exacerbated inflammation status: cytokines secreted by bone marrow cells, in particular megakaryocytes, probably activate endothelial cells and fibroblasts in the bone marrow microenvironment, contributing to the establishment of myelofibrosis.10

Elevated levels of cytokines have also been described in other hematological malignancies including chronic myeloid leukemia (CML), non-Hodgkin lymphoma, multiple myeloma and myelodysplastic syndromes.11–16 Moreover, the high plasmatic levels of inflammatory cytokines are associated with the response of CML patients to tyrosine-kinase inhibitors and with the prognosis of the aforementioned hematological malignancies.11–16

Considering the critical role that cytokines play in different hematological diseases and the contribution of inflammation to cancer development, the present investigation examined the differences between the cytokine profiles of ET, PMF and PV patients and analyzed their correlation with the patients’ clinical laboratory data.

Conventional statistical analyses were performed using GraphPad Prism 5.0 software (Graph-Pad Software, San Diego, CA, USA). Considering the nonparametric nature of all data sets, the Mann–Whitney test was used to compare the experimental groups, and the Spearman correlation test was used to analyze the correlation between cytokine/chemokine levels and the hematological parameters and clinical data. Statistical differences were considered significant when p-values were <0.05.

The correlations of cytokine/chemokine levels were compiled using the open source software Cytoscape (version 2.8; http://www.cytoscape.org).16 The networks were constructed using a circular layout, where the edges display the score of the correlation index (r) as negative (r<0), weak (r≤0.35), moderate (0.36≥r≤0.67) or strong (r≥0.68).18

Additionally, the cytokine profile of each individual was analyzed. First, all data from each cytokine/chemokine were used to calculate the overall median value, which was used as a cut-off point to classify each individual as a ‘high’ or ‘low’ producer of a given cytokine/chemokine. Next, the data were organized in black-and-white scale diagrams to calculate the frequency of high producers in each clinical group. Relevant data (≥50%), previously characterized by Luiza-Silva et al.19 were highlighted in both bold and underline format. Radar charts were built using Microsoft Excel (Microsoft Office 2013, Las Vegas, USA) to characterize the overall frequency of individuals expressing high levels of each cytokine/chemokine in each study group.

The analysis of overall cytokine profile was assessed to characterize the general cytokine pattern of each group as previously suggested by Vitelli-Avelar et al.20 but modified as follows: the percentages of cytokine/chemokine production were compiled using a four step platform, in this manner:

• (i)

  Conventional statistical analysis;

• (ii)

  Establishment of the overall median of cytokine/chemokine production for each group;

• (iii)

  Classification of each group as low or high cytokine-producer using the overall median values as the cut-off point;

• (iv)

  Creation of black-and-white scale diagrams for each group.

These approaches have been shown as relevant to detect, with high sensitivity, minor putative changes in the immunological profiles that are not detectable by conventional statistical approaches.19,21,22

The hallmarks of immune deregulation and onco-inflammation have been reported mostly in solid cancer patients, but also identified in some MPN patients.24 Increased levels of cytokines in PV and ET patients have indicated that they play a critical role in the physiopathology of MPN and so the measurement of their plasma levels might be useful for the clinical and therapeutic stratification of patients.25 This study assessed the different patterns of cytokine/chemokine expression in patients with three types of MPN (ET, PMF, and PV), their correlations with hematological parameters, associations with mutation status and interactions within networks.

Elevated plasma levels of several cytokines and chemokines were detected in ET, PMF and PV patients compared to healthy subjects. This finding indicates that an inflammatory process may be involved in the physiopathology of MPN, as cytokines and chemokines act in autocrine, paracrine and endocrine manners, and may even affect the hematopoietic niche. The levels of two regulatory cytokines, IL-4 and IL-10, were increased in all patient groups, which suggests an attempt to counterbalance the excessive pro-inflammatory response. In addition, the levels of cytokine/chemokine expression differed between the three MPN groups. For instance, PMF patients exhibited higher plasma levels of IFN-γ, IL-12p70 and IL-17A than ET patients and higher plasma levels of TNF-α, IL-4, IL12-p70 and GM-CSF than PV patients. ET patients exhibited higher plasma levels of RANTES than CTRLs and PV, as well as higher levels of MIP-1β than PV patients. IL-12 and IL-17 are pro-inflammatory cytokines that participate in both innate and adaptive immune responses and whose altered levels have been correlated to the physiopathology of various diseases, including cancer.26–28 RANTES, an important chemokine associated with cancer progression and metastasis, acts as a growth factor, stimulates angiogenesis and participates in immune evasion mechanisms.29,30 In this context, the increased levels of IL-12p70, IL-17A and RANTES in PMF patients suggest that they bear a more intense inflammatory process than ET or PV patients do.

The increased production of IP-10 in JAK2 V617F-positive PMF patients indicates a pro-inflammatory profile, as this cytokine, which is inducible by pro-inflammatory stimuli such as IFN-γ and TNF-α, presents pro-inflammatory properties.31 High levels of IP-10 seem to contribute to the pathogenesis of many organ-specific autoimmune diseases (such as autoimmune thyroiditis, type 1 diabetes, Graves’ disease and ophthalmopathy) and systemic autoimmune diseases (such as rheumatoid arthritis, systemic sclerosis, psoriatic arthritis, systemic lupus erythematosus and Sjögren syndrome), besides being an important chemokine involved in the alloimmune response against kidney allografts.32,33

The MPN driven JAK2 V617F mutation stimulates the JAK-STAT3/5 pathway to boost inflammatory cytokine production through autocrine and paracrine mechanisms.34 Positivity for the JAK2 V671F mutation is associated with leukocyte and platelet activation, and represents a thrombogenic factor in MPN.35,36 Little is known about the cytokine deregulation associated with thromboembolic events in MPN and their potential predictive values.10 Inflammation seems to be the major contributor to the prothrombotic state in MPN, as well as to bone marrow fibrosis, osteosclerosis, and angiogenesis.30 MPN patients also exhibit a massive upregulation of IFN-related genes, particularly the IFN-inducible gene IFI27, and severe deregulation of other inflammation and immune genes, as demonstrated by the whole-blood transcriptional profiling of several immune cells (granulocytes, monocytes, B cells, T cells, and platelets).37

Ho et al., using human cytokine array membranes combined to individual assets of enzyme-linked immunosorbent assay, enhanced chemiluminescence and high-throughput of microspot, evaluated plasmatic levels of 79 cytokines in controls and naïve PMF patients and described higher tissue inhibitor levels related to tissue inhibitor of matrix metalloproteinase-1 (TIMP-1), macrophage inflammatory protein-1β (MIP-1β), and insulin-like growth factor binding factor-2 (IGFBP-2).38 Differently from the current work they did not investigate these 79 cytokine plasmatic levels in PV and ET patients.39

Vaidya et al. showed by multivariable analysis and multiplex bead-based assaying that the elevated plasmatic levels of 13 cytokines (IL-1β, IL-4, EGF, IL-5, IL-7, IL-10, IL-17, TNF-α, IFN-α, GM-CSF, MIP-1β, MIP-1β and MCP-1) was associated with low overall survival in a group of 65 PV and 127 PMF patients. Moreover, similar to the current study, they showed that PMF and PV patients present different patterns of cytokine production.39

In the same context, Tefferi et al., using a multiplex biometric sandwich immunoassay and multivariable analysis reported increased levels of IL-1β, IL-1RA, IL-2R, IL-6, IL-8, IL-10, IL-12, IL-13, IL-15, TNF-α, G-CSF, IFN-α, MIP-1α, MIP-1β, HGF, IFN-γ–IP-10, monokines induced by IFN-γ (MIG), MCP-1 and vascular endothelial growth factor (VEGF) and decreased levels of IFN-γ in PMF patients compared to controls. The cytokine levels were associated with shorter survival in PMF patients.40

These authors studied a higher number of patients than the present study; however, they did not associate the patients’ mutation status and hematological parameters with the cytokine profile and did not compare the cytokine levels between the different MPN patient groups. The present study evaluated, 16 plasmatic cytokines/chemokines in ET, PMF, and PV patients at diagnosis using a multiplex bead-based assay to identify a disease signature/network and to associate it to the mutation status and hematological parameters. This study found that the studied MPN patients presented dysfunctional cytokine/chemokine production patterns characterized by high frequencies of ‘high’ cytokine/chemokine producers compared to healthy subjects. Furthermore, the correlation between the cytokine/chemokine levels and hematological parameters indicates that immune dysregulation may affect the production of hematological cells and favor their exacerbated proliferation.

Taken altogether, the data reported herein demonstrated that (i) the plasma cytokine/chemokine profile of ET, PMF, and PV patients are altered, which is most evident in PMF patients; and (ii) the JAK2 V617F status influences the cytokine/chemokine profile in PV patients. Our findings emphasize the overproduction of cytokines and chemokines in MPN patients, culminating in a pro-inflammatory state.

The increased levels of pro-inflammatory cytokines are associated with worse prognosis and shorter survival in MPN patients. Therapies targeting the pro-inflammatory cytokine profile seem to improve the quality of life and life expectancy of MPN patients. In this context, the JAK2 inhibitor ruxolutinib acts by decreasing the production of pro-inflammatory cytokines (inhibiting the cytokine gene transcription), thereby reducing the constitutional symptoms and splenomegaly (35%) in PMF patients.41,42 However, the JAK2 inhibitor potential against cytokine production in this therapy is not curative and does not impair MPN progression to leukemia. Thus, we must continue to study the molecular mechanisms involved in MPN pathogenesis to search for new efficient therapies for these disorders.

The authors declare no conflicts of interest.