Interleukin-6 (IL-6) is a pleiotropic cytokine that activates a classic signalling pathway upon binding to its membrane-bound receptor (IL-6R). Alternatively, IL-6 may ‘trans-signal’ in a manner that is facilitated by its binding to a soluble derivative of the IL-6 receptor (sIL-6R). Resultant signal transduction is, respectively, driven by the association of IL-6/IL-6R or IL-6/sIL-6R complex with the membrane-associated signal transducer, gp130 (Glycoprotein 130). Distinct JAK (Janus tyrosine kinase)/STAT (signal transducers and activators of transcription) and other signalling pathways are activated as a consequence. Of translational relevance, overexpression of IL-6 has been documented in several neoplastic disorders, including but not limited to colorectal, ovarian and breast cancer and several haematological malignancies. This review attempts to summarise our current understanding of the role of IL-6 in cancer development. In short, these studies have shown important roles for IL-6 signalling in tumour cell growth and survival, angiogenesis, immunomodulation of the tumour microenvironment, stromal cell activation, and ultimate disease progression. Given this background, we also consider the potential for therapeutic targeting of this system in cancer.
Interleukin-6 (IL-6) is a cytokine that has been known by a variety of names, reflecting its diverse actions on various cell types. Early studies identified this molecule as a T-cell derived stimulus to IgG production by B lymphoblastoid cell lines  and a B-cell growth factor . Following the independent cloning of the IL-6 gene in 1986, it became known as B-cell stimulatory factor-2 (BSF-2) or B-cell differentiation factor, reflecting its ability to promote the differentiation of B cells into immunoglobulin-producing cells . Interleukin-6 was also named interferon β2 (IFNβ2), due to its interferon-like antiviral activity . Identity of IL-6 with hepatocyte-stimulating factor was reported shortly thereafter, indicating its ability to promote the synthesis of acute phase proteins by hepatocytes . Finally, the myeloid blood cell differentiation-inducing protein type 2A (MGI-2A) proved to be identical to IL-6, reflecting the myeloid differentiating properties of this cytokine . To achieve a unified nomenclature for this pleiotropic cytokine, it was finally named as IL-6 in 1988 .
It is now appreciated that IL-6 is a multi-functional cytokine that plays important roles in inflammation [8,9], immune response , haematopoiesis [11,12], lipid metabolism , vascular disease , mitochondrial activity , insulin resistance , and in both the neuropsychological  and neuroendocrine systems . Moreover, it is now established that IL-6 is involved in several disease types, including autoimmune disease and cancer [19–23].
Structure of IL-6 and its receptor
Interleukin-6 is a member of the haematopoietic cytokine family and is a 21–28 kDa glycosylated protein composed of four long antiparallel α helices (A, B, C, and D) [24–26]. These helices are arranged in an up-up-down-down topology, forming three distinct epitopes which act as receptor-binding sites, namely sites I, II, and III [27–32] (Figure 1). Site I is formed by residues at the C-terminal of helix D and AB loop and acts as a binding site for the non-signalling IL-6 receptor subunit, IL-6Rα [31,33,34]. Site II is located on helix A and helix C and is formed by the binary interaction of IL-6 and IL-6Rα with the signal transducer glycoprotein (gp)130 between domain 2 and domain 3 [30,34]. Site III is composed of residues found in the N-terminal of the AB loop and residues at the C-terminal of helix D [32,35,36]. This site interacts with the Ig-like domain (IGD), also known as domain 1 (D1), of gp130 (Glycoprotein 130) to form a competent signalling complex [33,34,37–39].
Interleukin-6 receptor α is an 80 kDa glycoprotein, also named gp80 and CD126, that is expressed only on restricted cell lineages . This low-affinity receptor binds to IL-6 via residues within domain 2 and domain 3, resulting in the recruitment of the signal-transducing β-subunit gp130 to form a high-affinity IL-6R complex that is capable of signal transduction [40,41]. It is noteworthy that both the transmembrane domain and the cytoplasmic portion of IL-6R are not signalling competent, meaning that they can be removed without affecting IL-6 signalling . gp130, also called CD130 and IL-6Rβ, is a 130 kDa glycoprotein that is ubiquitously expressed in all tissues and acts as a signal transducer for the IL-6-type cytokine family, which includes interleukin-11 (IL-11), oncostatin M (OSM), leukaemia inhibitory factor (LIF), cardiotrophin-1 (CT-1), and ciliary neurotrophic factor (CNTF) . The gp130 subunit is unable to engage either IL-6 or IL-6Rα alone as it only associates with the complex IL-6/IL-6R through (i) the formed binary site II, which binds to the two cytokine-binding domains (CBDs) D2 and D3 of gp130 and (ii) site III to D1 of gp130 [34,43]. Functional studies have suggested that the high-affinity IL-6R complex is a hexamer, consisting of two distinct copies of IL-6, IL-6R, and gp130 [32,34]; however, a tetramer model containing one molecule of IL-6, IL-6R and two molecules of gp130 has also been proposed [44,45].
Signal transduction activated by IL-6
Once the IL-6/IL-6R/gp130 complex has assembled, multiple downstream signalling events take place that allow IL-6 to exert its pleiotropic actions, including differentiation, proliferation, survival, and apoptosis. Signals activated by the receptor complex are mediated via the Janus tyrosine kinase (JAK)–STAT (signal transducers and activators of transcription), Ras/Raf/mitogen-activated protein kinase (MAPK), phosphatidylinositol 3-kinase (PI3K), or Src/YAP pathways [46–52]. Upon gp130 dimerisation, JAK kinases interact through their FERM domain (Band-4.1, ezrin, radixin, moesin; also known as the N-terminal JAK-homology domain) with the membrane proximal box1 and box2 regions of gp130 [53,54]. This leads to JAK kinase activation and subsequent gp130 phosphorylation (Figure 2). Docking sites are created as a result that allows the recruitment and phosphorylation of STAT3, to a lesser extent STAT1 and to the least extent STAT5. Cells that lack JAK1 have impaired IL-6 signal transduction, indicating the crucial role of this particular kinase in IL-6 function . On the other hand, both JAK2 and Tyk2 (tyrosine kinase-2) appear to be dispensable in this regard . In turn, STAT3 can enhance the expression of IL-6, creating a feed-forward loop that has been implicated in cancer progression .
The activation of JAK–STAT pathway via IL-6.
JAK–STAT signalling via gp130 is strictly regulated through the involvement of various suppressors and inhibitors such as protein inhibitor of activated STAT (PIAS) proteins, suppressor of cytokine signalling proteins (SOCS), and cytokine-inducible SH2 (src homology-2) domain-containing protein (CISH), all of which function as feedback inhibitors that ultimately restrict IL-6 signalling . Illustrating this, mice that express a mutant gp130 lacking the ability to bind SOCS3 exhibit sustained activation of STAT1 and STAT3, resulting in the development of chronic inflammatory disease and cancer [59,60]. Further discussion of IL-6-mediated signal transduction is provided in a recent review .
Interleukin-6 receptor isoforms influence whether classic- or trans-signalling occurs
Interleukin-6 receptor exists not only as a membrane-bound receptor (mbIL-6R), but also as a soluble isoform (sIL-6R) . Soluble IL-6R is produced either via proteolysis of the ectodomain of the mbIL-6Rα by metalloproteinases including ADAM17 and ADAM10, or less frequently (1–20%) by alternative splicing of IL-6R mRNA [63–65]. Both forms of IL-6R bind IL-6 with similar affinity . Importantly, however, the complex formed between IL-6 and sIL-6R can initiate autocrine or paracrine IL-6 signalling in any cell type that expresses gp130, through a process known as IL-6 trans-signalling (Figure 3) . On the other hand, when IL-6 binds to a cell that expresses mbIL-6R, classic signalling ensues. Broadly speaking, trans-signalling promotes pro-inflammatory processes, while classic signalling is anti-inflammatory [68,69]. In both cases, gp130 homodimerisation is responsible for signalling. Recently, a putative third mode of IL-6 signalling was described by Heink et al. , referred to as ‘IL-6 trans-presentation’. In this mode of signalling, dendritic-associated membrane-anchored IL-6R binds to IL-6 and this complex engages gp130 on a recipient T cell. It should be noted that this system has been demonstrated using murine cells but remains untested in the human system.
Classic- and trans-signalling by IL-6.
To add complexity, gp130 also exists as many soluble isoforms. Soluble gp130 functions as a selective antagonist of IL-6 trans-signalling , binding exclusively to IL-6/sIL-6R to promote the formation of an inactive ternary complex . Importantly, sgp130 (soluble glycoprotein 130) is not believed to inhibit signalling by other cytokines that act via gp130 . Under normal circumstances, IL-6 is present at low picogram levels, considerably lower than either sIL-6R (∼75 ng/ml) or sgp130 (250–400 ng/ml). Consequently, Rose-John  has suggested that IL-6 is buffered by soluble factors that either promote (sIL-6R) or hinder (sIL-6R + sgp130) trans-signalling. Importantly, this buffering system may become dysregulated in inflammatory states in which circulating IL-6 levels can rise to achieve comparable levels to sIL-6R and sgp130 .
IL-6 lies at the heart of cancer-associated inflammation
In recent decades, several studies have linked chronic inflammation with the development of various malignancies, implicating IL-6 in this process [75,76]. Interleukin-6 is a crucial mediator of inflammation and can be found in substantial levels within the microenvironment of many cancers , including colorectal , ovarian , breast , prostate , lung , pancreatic , head and neck , and renal cell carcinomas , diverse lymphomas  and in multiple myeloma (MM) . In this context, IL-6 may be produced by tumour cells themselves, in addition to infiltrating stromal cells of mesenchymal and haematological origin (e.g. macrophages and T cells) . A key factor that drives IL-6 release in this context is the nuclear factor-kappa B (NF-κB) transcription factor, which is aberrantly hyperactivated in many cancers [88,89].
Within the tumour microenvironment, IL-6 can promote several pro-tumourigenic activities, including enhanced survival, growth, invasion and angiogenesis (described further below for individual cancer types). Additional effects of IL-6 include the activation of stromal fibroblasts , inhibition of type 1 immune responses , enhancement of the ratio of intra-tumoural regulatory to CD8+ T-cell types , enhanced generation of immunosuppressive myeloid cell subsets at the expense of dendritic cells [61,93], inhibition of natural killer cell function , and induction of epithelial-to-mesenchymal transition . Further discussion of the association of IL-6 with different immune cells are covered in the reviews [96,97].
Komoda et al.  showed that IL-6 levels in malignant tissues of 32 patients diagnosed with colorectal cancer (CRC) were significantly higher compared with normal colorectal tissues, concluding that IL-6 might have a role in tumour growth. Furthermore, Zeng et al.  reported that IL-6 can be a useful prognostic indicator since expression was significantly higher in CRC tissues compared with noncancerous tissues. Elevated serum levels of IL-6 have also been correlated with poorer disease outcome in CRC . In addition, increased IL-6 correlates with enhanced phosphorylation of STAT3 in CRC cells, but not in normal colon epithelium . The effect of IL-6 on CRC is most likely mediated through trans-signalling since Becker et al.  showed that sIL-6R was highly induced during tumorigenesis, while mbIL-6R was barely detected. This was further confirmed by detecting increased expression of ADAM17 on tumour cells compared with normal cells, an enzyme that is known to efficiently cleave mbIL-6R, releasing sIL-6R. More recently, Rose-John and colleagues have implicated IL-6 trans-signalling in a mouse model of colorectal cancer, acting downstream of epidermal growth factor signalling in myeloid cells .
Patients with advanced epithelial ovarian cancer (EOC) commonly have impaired immune function that is associated with elevated serum levels of pro-inflammatory molecules, including IL-6 . Several studies have implicated IL-6 in EOC progression , chemotherapy resistance , and poorer survival [107,108]. Levels of IL-6 are substantially elevated in the cystic fluid and ascites of advanced malignant compared with benign ovarian tumours and levels correlate with tumour size and volume of ascites [109–111].
Evidence suggests that IL-6 may play a direct role in facilitating ovarian tumour progression. One mechanism that may contribute to this is the up-regulated expression of anti-apoptotic molecules, including Bcl-2 (B-cell lymphoma 2) and Bcl-xl, that favour enhanced survival of EOC cells . Moreover, Wang et al.  demonstrated that IL-6 alters cell cycle distribution of ovarian cancer cells. Rabinovich et al.  demonstrated that IL-6 can also induce the secretion of matrix metalloproteinase-9 (MMP-9), augmenting the invasiveness of ovarian tumour cells. The addition of anti-IL-6 antibodies to SKOV-3 ovarian cancer cells resulted in reduced MMP-9 levels compared with control cultures.
Interleukin-6 has also been shown to act as a potent pro-angiogenic factor in different ovarian cancer models . In addition, increased production of secreted sIL-6R by ovarian cancer cells has been reported, most notably by SKOV-3 cells . This raises the possibility that IL-6 trans-signalling is involved in inducing tumour angiogenesis in ovarian cancers.
In vitro studies of the effects of IL-6 on breast cancer cell lines have yielded inconsistent results, demonstrating tumour-promoting activities in some models and tumour-suppressing actions in others. Illustrating this, IL-6 was detected in high quantities in supernatants of MCF-7/adriamycin multidrug-resistant breast cancer cells, whereas the cytokine was not detectable in the supernatants of the parental sensitive cells. Furthermore, the addition of exogenous IL-6 to MCF-7 cells caused an 8- to 10-fold increase in their resistance to doxorubicin treatment, while MCF-7 cells that had been engineered to produce IL-6 showed a 70-fold increase in their resistance to multiple drugs . The addition of exogenous IL-6 to MDA-MB-231 triple-negative breast cancer cells induced the production of G96, a protein related to drug resistance. In keeping with this, the authors found that both IL-6 and G96 proteins were considerably increased in breast tumours in comparison with non-malignant breast tissues . Nonetheless, other studies suggest that IL-6 may exert anti-tumour activity against some breast cancers. For example, IL-6 has been reported to induce apoptosis of MCF7 breast cancer cells in some studies [117,118]. In alignment with these findings, Shen et al.  demonstrated that along with IL-6, IL-1β and tumour necrosis factor (TNF)-α inhibited the proliferation of MCF-7 cells via suppression of insulin-like growth factor-I (IGF-I), a factor that is known to induce DNA synthesis in this cell type.
Regardless of its controversial role in breast cancer proliferation and survival, IL-6 can promote the migration of breast malignant cells, suggesting a role in metastasis [120,121]. The addition of IL-6-stimulated breast cancer cell spreading via induction of the transition from a stationary to motile phenotype . Another study indicated that IL-6 has anti-adhesive effects on breast cancer cells, a finding that was linked to reduced expression of E-cadherin by the tumour cells .
Although different studies have shown divergent results regarding the utility of IL-6 levels in breast tumour tissue as a prognostic factor [124–127], elevated serum IL-6 levels have been linked to negative prognosis in this disease. When compared with healthy women, serum IL-6 levels were significantly elevated in breast cancer patients and were correlated with the clinical stage of disease . Zhang and Adachi  demonstrated that patients with widely metastatic breast cancer had significantly increased serum IL-6 levels compared with patients with a single metastatic site only. They also reported that patients with higher levels of IL-6 had inferior response to chemo-endocrine therapy as well as reduced survival. Supporting these findings, other studies have linked the presence of increased levels of serum IL-6 in metastatic breast cancer patients with poorer progression-free and overall survival [130–132]. Levels of IL-6 were also notably higher in patients with recurrent compared with non-recurrent breast cancer. A similar increase was noted in patients with progressive recurrent disease in comparison with stable recurrent cases, indicating the potential use of IL-6 as a negative prognostic marker [133,134].
Other types of cancer
It is beyond the scope of this mini-review to provide a detailed summary of the role of IL-6 signalling in all types of cancer, although a brief overview is provided for completeness. In prostate cancer, similarly discordant findings to those obtained in breast cancer have been obtained with respect to the role of this cytokine in tumour cell apoptosis [135,136], although once again, IL-6 has also been linked to increased tumour cell invasiveness and metastasis [137–139].
Similarly, investigations have shown that IL-6 levels are increased in lung cancer patients, activating JAK2–STAT3 signalling pathway which in turn promotes tumour progression, drug resistance, and poor survival of patients [140–143].
In pancreatic cancer, serum IL-6 levels have been proposed as a prognostic marker while intra-tumoural activation of a JAK–STAT3 pathway was linked to particularly poor prognosis [144,145]. The ability of IL-6 to activate Pim-I kinase in pancreatic cancer has also been linked to enhanced tumour cell survival and treatment resistance [146,147]. The involvement of IL-6 trans-signalling was also described in promoting pancreatic ductal adenocarcinoma development .
Comparable findings were found in squamous cell carcinoma of head and neck where increased concentrations of IL-6 were correlated with advanced stage of disease . Duffy et al.  performed a study on 444 patients and concluded that serum IL-6 alone is a useful predictive marker of disease recurrence and poor prognosis. Tumour progression in this context has been linked to IL-6-mediated STAT3 activation .
In renal cell carcinoma (RCC), it was found that many cell lines constitutively secrete IL-6, resulting in enhanced proliferation and tumour progression. Furthermore, in the presence of a mutated TP53 (tumour protein-53) gene, IL-6 was secreted at greater levels . Activated STAT3 not only induces cell proliferation but also up-regulates the expression of vascular endothelial growth factor (VEGF) by activating hypoxia-induced factor-1 alpha (HIF-1α), thus promoting angiogenesis .
In haematological malignancies such as lymphoma and multiple myeloma, IL-6 is produced by the malignant cells themselves and bone marrow stromal cells. Moreover, IL-6 is produced by germinal centre B cells in Castleman's disease (CD) [153–155]. Similar to solid tumours, IL-6 regulates different signalling pathways that include Ras/Raf/MAPK pathway, PI3K/AKT pathway, and most importantly STAT3. In turn, this promotes cell survival and migratory phenotype, therapeutic resistance, angiogenesis, and metastasis [156–158]. Elevation of serum IL-6 and sIL-6R were also documented in non-Hodgkin's lymphomas (NHL), Hodgkin's disease (HD), and in adult T-cell leukaemia, potentially contributing to disease prognosis .
Targeting the IL-6/IL-6R axis in cancer
Since evidence implicates aberrant IL-6 production in the pathophysiology of various malignancies and chronic inflammatory diseases, inhibition of IL-6 signalling may provide a therapeutic opportunity in these disease types. This can be achieved by targeting the cytokine itself, IL-6 receptors, or the signalling pathways that are activated as a consequence. Many different strategies to achieve these goals have been developed in recent decades. In an early study, Klein et al. [155,160] demonstrated that a mouse anti-IL-6 monoclonal antibody (mAb) named B-E8 could inhibit malignant cell proliferation in patients with multiple myeloma. Furthermore, the mAb abrogated fever and normalised body temperature with no major adverse effects. Moreover, C-reactive protein (CRP), whose production is controlled by IL-6, was no longer detected in the circulation after treatment commencing, providing a potential marker for therapy efficacy. Through binding to IL-6, B-E8 prevents interaction with IL-6R and therefore inhibits active receptor complex assembly . A less immunogenic chimeric mAb with enhanced half-life called siltuximab (CNTO 328) was subsequently developed and has been evaluated in patients with MM, CD, NHL, prostate cancer, and renal cancer, with promising results in many cases [162–167]. In 2014, siltuximab was approved by the United States Food and Drug Administration (FDA) for the treatment of multicentric CD [168,169].
A strategy of blocking IL-6 signalling by preventing the formation of the active signalling complex has been achieved using human IgG1 Fc-tagged recombinant soluble gp130, namely olamkicept [72,73]. By binding to the IL-6/sIL-6R complex, it selectively inhibits trans-signalling that would otherwise arise following binding to membrane-bound gp130. In vivo testing has shown efficacy in different diseases including inflammation-induced cancer and Phase II trials are currently ongoing in this respect [68,72,73]. Moreover, there are two ongoing Phase II clinical trials of this blocking agent that are ongoing in patients with active inflammatory bowel disease [170,171].
Tocilizumab is a humanised mAb that blocks IL-6 signalling by interacting with both the membrane-bound and soluble forms of IL-6R, thereby preventing interaction with IL-6 [172,173]. This treatment has demonstrated significant efficacy in various diseases including CD  and rheumatoid arthritis , leading to FDA-approval for patients with rheumatoid arthritis and giant cell arteritis . Notably, tocilizumab has also been approved by the FDA for use in patients who develop cytokine release syndrome (CRS) following treatment with chimeric antigen receptor (CAR)-engineered T-cells  — a scenario in which macrophage-derived IL-6 signalling has been shown to be pivotal [178–180]. Several studies are ongoing to investigate the activity of tocilizumab in different cancer types.
Signal transduction downstream of IL-6 may also be antagonised using many approaches. Small molecule tyrosine kinase inhibitors have been designed that preferentially antagonise JAK1 and JAK3. Tofacitinib is such a molecule and has been approved by the FDA for the treatment of rheumatoid arthritis and ulcerative colitis . As for malignancies, Ruxolitinib, which selectively targets JAK1 and JAK2, was approved by the FDA for the treatment of myelofibrosis and polycythaemia vera . Alternatively, STAT3 inhibitors are currently undergoing clinical evaluation in many disease indications . The targeting of IL-6, its associated receptors and downstream pathways, is further discussed in several comprehensive reviews [97,182].
To date, most of the strategies used to block the IL-6/IL-6R axis have focused on inflammatory and myeloproliferative disorders rather than solid tumours, as indicated above. Therefore, more studies are needed to incorporate such agents with conventional treatment plans, including chemotherapy and radiotherapy, or with more recently developed therapies, such as immune checkpoint inhibitors. In fact, combination immunotherapy using anti-IL-6 antibody and PD-L1 blockade has shown a synergistic antitumour effect in murine models of hepatocellular carcinoma cancer, pancreatic cancer, and melanoma [91,183,184], providing a rationale for related clinical studies.
Overexpression of IL-6 has been identified in numerous types of cancer, leading to the dysregulation of a plethora of cellular activities that generally promote tumour progression. Strategies aimed at blocking IL-6 signalling may provide a potential therapeutic strategy for multiple malignancies, most likely as a component of a combinatorial therapeutic strategy. Further studies are needed to clarify the exact roles of classic versus trans-signalling in these actions of IL-6, since this may provide a rationale for more targeted inhibition of the function of this pleiotropic cytokine.
epithelial ovarian cancer
- FERM domain
Band-4.1, ezrin, radixin, moesin
interleukin-6 receptor alpha
Janus tyrosine kinase
mitogen-activated protein kinase
membrane-bound IL-6 receptor
soluble glycoprotein 130
soluble IL-6 receptor
suppressor of cytokine signalling
signal transducers and activators of transcription
Research in J.M.'s laboratory is supported by the Medical Research Council, Cancer Research UK, Bayer HealthCare, Bloodwise, the King's Health Partners Research and Development Fund, British Council Newton Fund Institutional Links Award, the British Lung Foundation, Breast Cancer Now, the Experimental Cancer Medicine Centre at King's College London, the King's Health Partners Cancer Research UK Cancer Centre and by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy's and St Thomas’ NHS Foundation Trust and King's College London.
The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. J.M. is chief scientific officer of Leucid Bio, a spinout company focused on CAR T-cell and Vγ9Vδ2 T-cell immunotherapies for malignant disease. M.Y.T. was sponsored by the Saudi Arabian Cultural Bureau.
The Authors declare that there are no competing interests associated with the manuscript.