The emerging roles of inflammasome‐dependent cytokines in cancer development

Abstract

In addition to the genomic alterations that occur in malignant cells, the immune system is increasingly appreciated as a critical axis that regulates the rise of neoplasms and the development of primary tumours and metastases. The interaction between inflammatory cell infiltrates and stromal cells in the tumour microenvironment is complex, with inflammation playing both pro‐ and anti‐tumorigenic roles. Inflammasomes are intracellular multi‐protein complexes that act as key signalling hubs of the innate immune system. They respond to cellular stress and trauma by promoting activation of caspase‐1, a protease that induces a pro‐inflammatory cell death mode termed pyroptosis along with the maturation and secretion of the pro‐inflammatory cytokines interleukin (IL)‐1β and IL‐18. Here, we will briefly introduce inflammasome biology with a focus on the dual roles of inflammasome‐produced cytokines in cancer development. Despite emerging insight that inflammasomes may promote and suppress cancer development according to the tumour stage and the tumour microenvironment, much remains to be uncovered. Further exploration of inflammasome biology in tumorigenesis should enable the development of novel immunotherapies for cancer patients.

Glossary

3‐MCA

3‐methylcholanthrene

AID

autoinflammatory diseases

AIM2

absent in melanoma 2

ALR

HIN200/AIM2‐like receptor

AML

acute myeloid leukaemia

AOM

azoxymethane

ASC

apoptosis‐associated speck‐like protein containing a CARD

CANTOS

Canakinumab Anti‐inflammatory Thrombosis Outcome Study

CAPS

cryopyrin‐associated periodic syndrome

CAR

chimeric antigen receptor

CARD

caspase recruitment domain

CRP

C‐reactive protein

DAMP

danger‐associated molecular patterns

dsDNA

double‐stranded DNA

DSS

dextran sodium sulphate

FDA

Food and Drug Administration

FMF

familial Mediterranean fever

GSDMD

gasdermin D

HMGB1

high‐mobility group box 1 protein

IBD

inflammatory bowel disease

IFN‐γ

interferon‐γ

IGIF

IFN‐γ‐inducing factor

IL‐18BP

IL‐18 binding protein

IL‐1RAcP

IL‐1 receptor accessory protein

IL‐1Ra

IL‐1 receptor antagonist

IL

interleukin

IRF8

interferon regulatory factor 8

LLC

Lewis lung carcinoma

LPS

lipopolysaccharide

MDSC

myeloid‐derived suppressor cell

MM

multiple myeloma

NAIP

NLR family apoptosis inhibitory protein

NK

natural killer

NLR

Nod‐like receptor

NSAID

non‐steroidal anti‐inflammatory drug

PAMP

pathogen‐associated molecular patterns

PMN

polymorphonuclear

SIRS

systemic inflammatory response syndrome

T3SS

type III secretion system

TGF‐β

transforming growth factor β

TLR

Toll‐like receptor

TNF

tumour necrosis factor

TRIM

tripartite motif

VEGF

vascular endothelial growth factor

Introduction

Cancer has long been considered as a cell‐autonomous process that depends on genetic alterations within malignant cells. However, it is now accepted that for a tumour to emerge, develop and invade a tissue, it requires close interactions not only between proliferating malignant cells, but also with stromal cells in its microenvironment (fibroblasts, endothelial cells), and with resident and infiltrating innate and adaptive immune cells. Indeed, Rudolf Virchow suggested a potential link between inflammation and cancer after observing inflammatory cell infiltrates in solid tumours over 150 years ago 1. Ever since, it has become evident that the immune system not only eradicates tumours by surveilling tissues and organs for neoplastic cells and neo‐antigens, but also is capable of supporting and shaping emergent tumours. Different lines of evidence indicate that chronic inflammation caused by persistent infections, autoimmune diseases and prolonged exposure to environmental irritants and dietary factors may contribute to tumorigenesis by creating a local environment that supports oncogenic mutations and genomic instability, and enhances angiogenesis among several other mechanisms that support the development and growth of tumours. Well‐established examples of infection‐sensitized cancer development are gastric cancers and mucosa‐associated lymphoid tissue lymphomas that are associated with persistent Helicobacter pylori infection 2, 3. Additionally, hepatitis B and hepatitis C virus infection provide an increased risk for the development of hepatocellular carcinoma and non‐Hodgkin lymphoma 3. Not only inflammation in the context of chronic infections, but also non‐communicable autoimmune diseases such as inflammatory bowel diseases (IBD) increase the risk for the development of colorectal cancer 4. In support of this notion, an important body of epidemiological findings and several randomized controlled trials demonstrate that suppressing chronic inflammation by the long‐term use of aspirin and other non‐steroidal anti‐inflammatory drugs (NSAIDs) significantly reduces the risk for colorectal, gastric, lung and other cancers 5, 6, 7.

Apart from the widely appreciated role chronic inflammation plays in promoting cancer development, smouldering inflammation can occur in the local tumour microenvironment, and this process is increasingly thought to exert potent tumour‐promoting effects. Tumour‐elicited inflammation can result in cancer cell killing, but this process may also promote survival and proliferation of malignant cells that successfully evaded destruction by antigen‐specific immune cells. In addition, local inflammatory responses may stimulate tumour development and metastasis by supporting local immunosuppression and subversion of antigen‐specific adaptive immune responses and by promoting the formation of a neovasculature that supports angiogenesis and metastasis 8. A range of cancer immunotherapeutic approaches that include monoclonal antibodies, immune checkpoint inhibitors, cancer vaccines and chimeric antigen receptor (CAR) T‐cell therapies aims to revive the patient's suppressed immune system to help to eradicate the disease. However, in addition to their direct effects on malignant cells, chemo‐ and radiotherapeutic agents may induce an inflammatory response that stimulates tumour re‐emergence and resistance to therapy 9, 10. Here, we will introduce inflammasome biology and discuss how inflammasome‐produced cytokines modulate cancer development.

IL‐1β, IL‐18 and inflammasome signalling

Cytokines are soluble immunomodulating proteins that are expressed on the plasma membrane and/or secreted in the extracellular environment. They act in autocrine and paracrine manners by binding on their cognate receptors on effector cells to promote or inhibit tumour development and progression. Examples of key cytokines that are implicated in tumorigenesis include tumour necrosis factor (TNF), interleukin‐6 (IL‐6), transforming growth factor‐ß (TGF‐ß) and vascular endothelial growth factor (VEGF). IL‐1β and IL‐18 are two additional cytokines that are rapidly emerging as central modulators of tumorigenic processes that may either promote or suppress tumour growth depending on the tumour type, stage and microenvironment. Unlike most other cytokines, IL‐1β and IL‐18 are produced as biologically inert pro‐cytokines that reside in the cytosol of naïve immune cells. The assembly of multi‐protein complexes termed inflammasomes activates the protease caspase‐1, which under most conditions is essential for the proteolytic maturation of proIL‐1β and proIL‐18 into highly inflammatory, secreted cytokines 11. Before discussing the dual roles of IL‐1β and IL‐18 in cancer development, we will briefly introduce the inflammasome signalling pathways that regulate their production.

Having been described for the first time by Tschopp and colleagues in 2002 12, inflammasomes are now considered central signalling hubs of the immune system 13. A substantial body of evidence demonstrates that these innate immune pathways are essential for protecting the host from bacterial, viral, fungal and protozoal infections and to cope with cellular stress 11. When deregulated, however, inflammasome signalling may contribute to a suite of autoinflammatory, autoimmune, neurodegenerative and metabolic diseases. Inflammasomes are defined as caspase‐1‐activating multi‐protein complexes that are assembled in response to infections, pathogen‐associated molecular patterns (PAMPs) and cellular stress that is detected through the concomitant production or delocalization of danger‐associated molecular patterns (DAMPs). Although the downstream effectors of the different inflammasome types are shared, specificity in their responses is provided by sensor proteins that directly or indirectly detect a suite of PAMPs and DAMPs, followed by their oligomerization, the recruitment of the bipartite inflammasome adaptor protein ASC and activation of caspase‐1 in the complex (Fig 1). In addition to promoting the maturation and extracellular release of the pro‐inflammatory cytokines IL‐1β and IL‐18, the induction of an inflammatory cell death mode termed pyroptosis represents another major physiological outcome of inflammasome activation. The ability of pyroptosis to promote inflammation resides in its lytic nature. It results from inflammatory caspase‐mediated proteolytic maturation of gasdermin D (GSDMD), the freed amino‐terminal domain of which oligomerizes and perforates the plasma membrane to promote extracellular release of mature IL‐1β and IL‐18, as well as DAMPs such as IL‐1α and high‐mobility group box 1 (HMGB1) 11, 13, 14, 15. Hence, recent studies have implicated GSDMD in familial Mediterranean fever (FMF) 16, neonatal‐onset multi‐system inflammatory disease 17, non‐alcoholic steatohepatitis 18 and a mouse model of multiple sclerosis 19.

Figure 1. General concept of inflammasome signalling mechanisms.

The canonical inflammasome pathway is triggered by multiple pathogens and inflammatory agents. The presence of PAMPs and DAMPs is sensed by specific innate immune sensors which results in their oligomerization and ultimately inflammasome assembly. Innate immune sensors known to engage well‐defined inflammasomes are a subset of NOD‐like receptors (NLRP1b, NLRC4 and NLRP3), a member of the HIN200/AIM2‐like receptor family (AIM2) and a member of TRIM family (Pyrin/TRIM20). The NLRP1b inflammasome responds to lethal factor produced by Bacillus anthracis and assembles an inflammasome by recruiting caspase‐1 through its CARD domain or through ASC as an adaptor. NAIPs function as direct receptors for bacterial flagellin and the type 3 secretion system (T3SS). Ligand binding results in NAIP activation, allowing it to recruit and activate the NLRC4 inflammasome which can form complexes with or without recruiting ASC. Upon cytosolic recognition of viral, bacterial or host‐derived double‐stranded DNA, the AIM2 inflammasome is assembled. Activation of the Pyrin inflammasome on the other hand is triggered by RhoA GTPase‐inactivating modifications induced by various bacterial toxins. Finally, the NLRP3 inflammasome responds to intracellular damage induced by pathogenic or sterile insults. Potassium efflux is thought to be the common event associated with the diverse array of NLRP3 stimuli. While the PYD‐based sensors (NLRP3, AIM2 and Pyrin) require the bipartite adaptor protein ASC to recruit caspase‐1, the CARD‐based sensors (NLRP1b and NLRC4) can recruit caspase‐1 independently of ASC. Oligomerization of pro‐caspase‐1 leads to its autoactivation, and active caspase‐1 catalyses the maturation and secretion of the inflammatory cytokines IL‐1β and IL‐18. Additionally, active caspase‐1 engages pyroptosis by cleaving its substrate gasdermin D. This lytic mode of programmed cell death coincides with the release of alarmins such as IL‐1α and HMGB1. The non‐canonical inflammasome pathway is activated when cytosolic LPS is sensed by pro‐caspase‐11 in mice (or pro‐caspase‐4/5 in humans). Upon oligomerization, active caspase‐11 cleaves gasdermin D to drive pyroptosis. In addition, active caspase‐11 induces non‐canonical activation of NLRP3 resulting in caspase‐1 activation and the maturation of the inflammatory cytokines IL‐1β and IL‐18, which are then released by gasdermin D‐induced pyroptosis. AIM2, absent in melanoma 2; ASC, apoptosis‐associated speck‐like protein containing a CARD; CARD, caspase recruitment domain; DAMP, damage‐associated molecular pattern; dsDNA, double‐stranded DNA; HMGB1, high‐mobility group box 1; LPS, lipopolysaccharide; NAIPs, NLR family apoptosis inhibitory proteins; NLR, NOD‐like receptor; NOD, nucleotide‐binding oligomerization domain; PAMP, pathogen‐associated molecular pattern; PYD, Pyrin domain; TRIM, tripartite motif.

Inflammasome types

To date, at least five distinct inflammasome types that are typically named after the intracellular sensor that assembles the complex have been validated genetically in mouse models (Fig 1). Several members of the nucleotide‐binding domain and leucine‐rich repeat‐containing (NLR) receptor family, when responding to intracellular danger, initiate the formation of an inflammasome complex. AIM2 (absent in melanoma 2), a member of the HIN200/AIM2‐like receptor (ALR) family, also is an inflammasome sensor 20, 21, 22, 23, 24. More recently, Pyrin/TRIM20, a member of the tripartite motif (TRIM) family of intracellular sensors, was shown to assemble a functional inflammasome 25. While in humans only a single NLRP1 gene exists, mice encode three paralogs: Nlrp1a, Nlrp1b and Nlrp1c 26. Bacillus anthracis lethal toxin is the only defined biochemical agent that activates the Nlrp1b inflammasome, whereas specific agents that activate the human NLRP1, and murine Nlrp1a and Nlrp1c inflammasomes await discovery 26. Uniquely, the NLRP3 inflammasome requires a two‐step mechanism for activation and responds to a large suite of activators and insults. Firstly, Toll‐like receptor (TLR)‐mediated priming provides for nuclear factor‐kappa B (NF‐κB)‐mediated transcriptional upregulation of NLRP3 and proIL‐1β 27. This sets the scene for NLRP3 activation through incompletely understood mechanisms following subsequent exposure to pore‐forming agents, crystals, β‐amyloids and many other NLRP3‐activating agents. Indeed, DAMPs such as extracellular ATP and hyaluronic acid; medically relevant crystalline products such as alum, CCPD, MSU, silica and asbestos; ionophores such as nigericin; and β‐fibrils can all trigger assembly of the NLRP3 inflammasome 28, 29, 30, 31. Basal activity of the serine/threonine kinase TAK1 was recently shown to restrain NLRP3 inflammasome activation, and pathogen blockade of TAK1 promotes caspase‐8‐dependent cleavage of GSDMD and NLRP3 inflammasome‐dependent release of IL‐1β 32, 33. Moreover, the major component of the outer membrane of Gram‐negative bacteria, lipopolysaccharide (LPS), was shown to activate NLRP3 through a non‐canonical pathway involving caspase‐11. Upon detection of intracellular LPS, caspase‐11 autonomously induces pyroptosis and through the NLRP3 inflammasome triggers caspase‐1‐dependent IL‐1β and IL‐18 maturation 34, 35. Intracellular detection of bacterial flagellin or the type III secretion system (T3SS) of, e.g., Salmonella Typhimurium results in activation of the NLRC4 inflammasome. Cytosolic recognition of these bacterial factors by members of the NLR family apoptosis inhibitory protein (NAIP) cluster within the NLR family mediates NLRC4 inflammasome activation 36, 37. Constitutive expression of NAIP family members is under control of interferon regulatory factor 8 (IRF8), which thus is required for optimal activation of the NLRC4 inflammasome 38. Upon cytosolic recognition of viral (e.g. vaccinia virus), bacterial (e.g. Francisella tularensis and Listeria monocytogenes) or host‐derived double‐stranded DNA (dsDNA), the AIM2 inflammasome is assembled 23, 39, 40, 41, 42. Finally, it was recently established that RhoA GTPase‐inactivating modifications induced by various bacterial toxins (e.g. Clostridium difficile) trigger activation of the Pyrin inflammasome 25.

Inflammasome assembly and caspase‐1 activation

Following the detection of PAMPs and DAMPs by the respective inflammasome sensors, the inflammasome adaptor apoptosis‐associated speck‐like protein containing a CARD (ASC) is recruited. This adaptor protein bridges interactions between inflammasome sensors and caspase‐1, and promotes caspase‐1 oligomerization and the formation of a single micrometre‐sized supramolecular fibril structure named the “ASC speck”. ASC is essential for the Pyrin (PYD) domain‐containing sensors (NLRP3, AIM2 and Pyrin) to recruit caspase‐1, while the caspase recruitment domain (CARD)‐based sensors (NLRP1b and NLRC4) may also recruit caspase‐1 directly as shown by the induction of caspase‐1‐mediated pyroptosis and cytokine secretion from ASC‐deficient macrophages 43, 44. Caspase‐1 is expressed as a cytosolic inactive zymogen, but gains proteolytic activity through proximity‐induced conformational changes imposed by its recruitment in inflammasome complexes. Under physiological conditions, this results in caspase‐1 autocleavage and enhanced cleavage of IL‐1β and IL‐18, although studies have shown that caspase‐1 autocleavage is dispensable for caspase‐1 protease activity 43, 44.

Inflammasome‐mediated maturation of IL‐1 family cytokines IL‐1β and IL‐18

Akin to caspase‐1, IL‐1β and IL‐18 are stored as inactive proforms in the cytoplasm. In contrast to unprocessed IL‐1α that is constitutively active, caspase‐1‐dependent cleavage of IL‐1β and IL‐18 is required for gaining biological activity 45. Caspase‐1 processes IL‐1β into a 17 kDa mature fragment 46. Similarly, caspase‐1‐mediated proteolytic processing of IL‐18 results in a mature product of 17.2 kDa that is released extracellularly 47, 48. When produced locally in inflamed tissues, IL‐1β mainly functions to stimulate leucocyte activation. Because systemic IL‐1β is a prominent inducer of fever and acute‐phase proteins that at high levels may even trigger systemic inflammatory response syndrome (SIRS), its levels in circulation must be tightly regulated. Indeed, IL‐1β is hardly detected in serum, which is probably due to its short half‐life as well as its rapid neutralization by soluble receptors 49, 50, 51. Additionally, responsiveness to IL‐1β is limited by IL‐1 receptor antagonist (IL‐1Ra) and surface‐bound IL‐1RII that both prevent IL‐1‐induced signalling by respectively preventing bio‐active IL‐1β from binding to its receptor and functioning as a decoy receptor. Constitutively, active IL‐1α signals through the same receptor as IL‐1β, and collectively, they represent the pro‐inflammatory IL‐1 cytokine members. In this review, we will focus on IL‐1β, although the reader should keep in mind that IL‐1α uses the same receptor and signalling mechanisms to induce inflammatory responses. IL‐18, on the other hand, is readily detectable in serum, and levels appear to correlate with age in apparently healthy individuals 52. Similar to IL‐1β, IL‐18 is sequestered and neutralized in circulation, in this case by soluble IL‐18 binding protein (IL‐18BP) 53 (Fig 2).

Figure 2. IL‐1β and IL‐18 signalling—receptors, decoys and natural antagonists.

(A) Both IL‐1α and IL‐1β bind to IL‐1R1, thereby enabling recruitment of the IL‐1RAcP coreceptor. Approximation of the intracellular TIR domains of the IL‐1R complex results in recruitment of MyD88 followed by a cascade of downstream events to ultimately result in the activation of important signalling proteins, such as mitogen‐activated kinases (e.g. p38), as well as transcription factors, including NF‐κB, which control expression of a number of inflammatory genes. Besides agonistic molecules leading to an inflammatory response, several receptors, co‐receptors, antagonists and decoy receptors that dampen immune signalling by IL‐1 family cytokines are known. Membrane‐bound IL‐1R2 binds with high affinity to IL‐1α and IL‐1β, but because of the lack of a TIR domain, downstream signalling is blocked. Soluble IL‐1Ra functions as a receptor antagonist by binding with similar affinity to the receptor and blocking its signalling. Furthermore, signalling through the IL‐1R complex is modulated by soluble forms of the receptors (sIL‐1R1, sIL‐1R2 and sIL‐1RAcP). (B) Upon binding of IL‐18 to IL‐18Rα, the latter one heterodimerizes with IL‐18Rβ. Similar to the IL‐1R complex, approximation of the 2 receptors results in recruitment of MyD88 and a downstream cascade activating signalling proteins and transcription factors. Signalling through the IL‐18R complex is modulated by inhibitory actions of IL‐18BP that binds and neutralizes mature IL‐18 in circulation. IL‐18BP, IL‐18 binding protein; IL‐1R, IL‐1 receptor; IL‐1Ra, IL‐1 receptor antagonist; IL‐1RAcP, IL‐1 receptor accessory protein; TIR, Toll/IL‐1 receptor.

Cytokine release following inflammasome activation is vital in sustaining homeostasis, but when signalling is deregulated, aberrant IL‐1β and IL‐18 production may become detrimental to the host, as shown by the extensive catalogue of human diseases in which these cytokines contribute to pathology. This is pertinent in a suite of autoimmune and autoinflammatory diseases. Some autoinflammatory diseases (AID) are due to gain‐of‐function mutations in key inflammasome signalling molecules, leading to periodic fevers. Examples include cryopyrin‐associated periodic syndrome (CAPS), FMF and NLRC4‐AID 54. Being caused by deregulated inflammasome signalling, patients suffering from these periodic fever syndromes benefit highly from IL‐1‐ and IL‐18‐neutralizing therapies. Increasing evidence suggests that more common diseases such as gout, type 2 diabetes, atherosclerosis, recurrent pericarditis and osteoarthritis, and lung cancer also are responsive to IL‐1β neutralization 55, 56. Although the underlying mechanisms and causal connections remain unclear in most cases, there is a steady increase in the number of clinical studies showing the efficacy of anti‐IL‐1 and anti‐IL‐18 therapies in different diseases. In the following sections, we will focus on discussing the emerging roles of inflammasome‐released cytokines in cancer and the therapeutic potential of IL‐1β and IL‐18 neutralization in prevention, interception and cure of cancer.

The dual roles of IL‐1β in cancer

While not being expressed under homeostatic conditions, the levels of IL‐1β are elevated in numerous tumours, including melanoma, colon and breast cancer as well as in haematological malignancies 57, 58, 59, 60, 61, 62, 63. Notably, clinical studies revealed that patients with IL‐1β‐producing tumours generally have poorer prognosis 56, 59, 62, 64, 65. Cancer cells can directly produce IL‐1β due to intrinsic alterations in cytokine gene expression patterns or in response to host‐derived cytokines, and they can elicit IL‐1β production from stromal cells and infiltrating leucocytes 57, 66. As a pleiotropic cytokine, IL‐1β influences many cell types. In addition to being one of the most pyrogenic molecules in the human body, it can increase vascular permeability and expression of adhesion molecules, matrix metalloproteinases and other cytokines, while also triggering the chemotactic recruitment of granulocytes. These activities are beneficial to the host in the context of microbial infections as increased adhesion molecule expression on the endothelium facilitates the emigration of neutrophils into the tissues, which contribute to killing of microbial pathogens. However, these same IL‐1β‐mediated effector mechanisms may facilitate tumorigenesis, metastasis of tumours and other harmful activities in solid tumours and haematological malignancies (Fig 3).

Figure 3. Overview of IL‐1β and IL‐18 effector mechanisms in cancer.

Both IL‐1β and IL‐18 may be expressed at high levels in tumour tissues, where they may originate from tumour, stromal and immune cells. Both cytokines can either suppress or promote tumorigenesis depending on the tumour type, stage and the tumour microenvironment. IL‐1β and IL‐18 may act to suppress anti‐tumour immunity. Additional pro‐tumorigenic mechanisms include the induction of angiogenic factors, adhesion molecules, matrix metalloproteinases, inflammatory cytokines and the recruitment of circulating immunosuppressive cells. Through their pleiotropic actions, IL‐1β and IL‐18 may suppress or augment tumour initiation, growth, invasion and metastasis. DC, dendritic cell; NK, natural killer; MMP, matrix metalloproteinases; MDSC, myeloid‐derived suppressor cell.

Pro‐tumorigenic role of IL‐1β in solid tumours

Tumour development can be divided into different stages going from tumour initiation, over growth to invasion and metastasis. While there is ample evidence for a role of IL‐1β in tumour progression and metastasis, less is known about its contribution to the initial step of tumour initiation 57. However, a mouse model of 3‐methylcholanthrene (3‐MCA)‐induced skin carcinogenesis provided evidence for significantly impaired tumour development in IL‐1β‐deficient mice with only 40% of the challenged animals developing tumours, while contrastingly tumorigenesis was markedly accelerated in IL‐1Ra‐deficient mice 67. Additional findings supporting a role for IL‐1β in the induction of neoplasia were provided by experiments with mice that overexpress human IL‐1β in the stomach, which caused the spontaneous and stepwise development of inflammation, metaplasia, dysplasia and carcinoma of the stomach, respectively. In addition, Helicobacter felis infection in these animals caused accelerated progression to gastric atrophy and cancer 68. Together, these observations add support to the suggested clinical association between IL‐1β single nucleotide polymorphisms that are thought to increase IL‐1β production and the risk of gastric cancer in Helicobacter pylori‐infected patients 69.

Due to its pleiotropic nature, several mechanisms of how IL‐1β may promote metastasis have been put forward 56. Almost 30 years ago, it was shown that a single injection of recombinant human IL‐1β prior to administration of tumour cells increased experimental lung metastases in nude mice 70. The proposed underlying mechanism was IL‐1β‐induced stimulation of endothelial adhesion molecules, which enhanced adherence of tumour cells. Further support for a critical role of IL‐1β in tumour cell metastasis emerged from studies in IL‐1β‐deficient mice that showed a 90% reduction of hepatic metastasis of mouse B16 melanoma cells that had been injected in the spleen 71. This was consistent with previous findings showing that intravenous injection of recombinant IL‐1β enhanced hepatic metastasis of intrasplenically injected B16 cells 72. Interestingly, IL‐1β‐stimulated B16 metastasis appears tissue‐specific because treatment with recombinant human IL‐1Ra reduced B16 metastasis in bone marrow, spleen, liver, lung, pancreas, skeletal muscle, adrenal gland and heart, while failing to reduce metastases to the kidney, testis, brain, skin and the gastrointestinal tract 73. IL‐1β also is a potent inducer of angiogenesis, a process that supports progressive growth of malignant tumours by supplying oxygen and nutrients to the tumour cells through the formation of new blood vessels 74, 75. In one of the earlier studies, murine Lewis lung carcinoma (LLC) cells expressing human IL‐1β grew rapidly in mice compared to wild‐type LLC cells. This rapid growth was associated with increased neovascularization induced by several angiogenic factors (including VEGF, CXCL2 (mouse functional homolog of human IL‐8) and hepatocyte growth factor) that were secreted by both tumour cells and stromal cells in the tumour microenvironment 76. Additional evidence that supports a role for IL‐1β in angiogenesis and metastasis came from in vivo studies showing reduced local tumour growth and metastasis in IL‐1β‐deficient mice compared to wild‐type mice in two different cancer models, namely the B16 melanoma and the DA/3 mammary adenocarcinoma model. Furthermore, Matrigel plugs were used to assess the recruitment of blood vessel networks for B16 melanoma cells. Vascularization of the plugs was observed in wild‐type mice, but absent in IL‐1β‐deficient mice. Recombinant IL‐1Ra inhibited the growth of blood vessel networks into B16‐containing Matrigel plugs in wild‐type mice 77. In another study, neutralization of IL‐1β abrogated cell infiltration and angiogenesis in Matrigel plugs that were implanted in mice 78. When explanted, the Matrigel plugs contained 85% less VEGF. Similarly, supernatants of IL‐1β‐deficient macrophages did not induce inflammatory or angiogenic responses. Other studies showed that IL‐1β upregulates HIF‐1α and VEGF, further confirming a key role for IL‐1β in angiogenesis 79.

Apart from actively promoting tumour growth, IL‐1β also promotes tumour development by compromising anti‐cancer immunity. This is caused by the recruitment of myeloid‐derived suppressor cells (MDSCs), which are a heterogeneous population of immature myeloid cells 80, 81. Xenograft tumours that overexpress IL‐1β showed greater accumulation of MDSC and more rapid tumour progression compared to control tumours. Notably, resection of large tumours of IL‐1β‐secreting cells restored immune reactivity within 7–10 days. Treatment of tumour‐bearing mice with IL‐1Ra also reduced tumour growth 82. Similar findings were obtained with 4T1 mammary carcinoma tumours. When implanted into IL‐1R1‐deficient mice, these tumours exhibited delayed accumulation of MDSC and slower growth 83. Furthermore, stomach‐specific expression of human IL‐1β in transgenic mice led to spontaneous gastric inflammation and cancer that correlated with early recruitment of MDSC to the stomach. Antagonism of IL‐1 receptor signalling by recombinant IL‐1Ra suppressed MDSC mobilization and inhibited the development of gastric preneoplasia, further supporting the role of IL‐1β in MDSC recruitment and carcinogenesis 68.

Anti‐tumorigenic role of IL‐1β in solid tumours

Although tumour‐promoting effects of IL‐1β are well‐documented, in particular during cancer metastasis and tumour angiogenesis, there is also evidence for a cancer‐suppressive effect of this pro‐inflammatory cytokine 84, 85. Already about 30 years ago has it been shown that intraperitoneal injection of recombinant human IL‐1β caused complete regression of subcutaneous SA1 sarcoma and L5178Y lymphomas in mice 86. While causing regression when dosed on days 6–8 of tumour growth, recombinant IL‐1β had no effect when given on days 1–3, suggesting that the anti‐tumour action of IL‐1β is based on an underlying host‐immune response that is not generated until after day 3 of tumour growth. Direct evidence for the participation of host immunity in IL‐1β‐induced tumour regression was provided by results showing that IL‐1β lost its tumour‐suppressive effect in T‐cell‐deficient mice, indicating that IL‐1β functions by stimulating T‐cell‐mediated anti‐tumour immunity 86. More recently, this finding was corroborated and further developed in a study showing that myeloma eradication by Th1 cells was not affected by inhibition of TNF, TNF‐related weak inducer of apoptosis (TWEAK) or TNF‐related apoptosis‐inducing ligand (TRAIL), but was, however, severely impaired by the in vivo neutralization of IL‐1β (as well as IL‐1α) with recombinant IL‐1Ra 87. Not only the anti‐tumour functions of tumour‐specific Th1 cells, but also those of tumour‐infiltrating macrophages were affected by IL‐1 neutralization 87. Moreover, IL‐1β‐producing dendritic cells appear to play a critical role in the therapeutic efficacy of anti‐cancer chemotherapy. Dendritic cells cross‐present neo‐antigens from dying cancer cells to prime tumour‐specific interferon‐γ (IFN‐γ)‐producing T lymphocytes. Priming of these IFN‐γ‐producing CD8+ T cells by dying tumour cells, however, fails in the absence of functional IL‐1 receptor signalling and in mice that are deficient in the inflammasome components Nlrp3 and caspase‐1 unless exogenous IL‐1β is supplied. Collectively, these results imply that inflammasome signalling ameliorates T‐cell‐mediated immunity against tumour cells 88.

Pro‐tumorigenic role of IL‐1β in haematological malignancies

IL‐1β is not only pro‐tumorigenic in solid tumours, but also drives progression of haematological malignancies, exemplified by acute myeloid leukaemia (AML) and the plasma cell malignancy multiple myeloma (MM), respectively 58. It is now well‐appreciated that hematopoietic stem and progenitor cells acquire somatic mutations with ageing, resulting in the emergence of (sub)clonal haematopoiesis and the development of overt haematological malignancies 89. Recently, cell‐extrinsic factors such as infection and inflammation have been put forward as potential mechanisms that promote a selective competitive advantage to mutant cells to expand and survive 58, 89, 90. Furthermore, aberrant innate immune and cytokine signalling, as well as perturbed anti‐tumour immunity, seem to be pivotal in the pathogenesis of haematological malignancies 58.

Acute myeloid leukaemia is a genetically heterogeneous disease characterized by clonal expansion of myeloid progenitors that accumulate in the bone marrow and peripheral blood. Given the diverse genetic and molecular abnormalities underlying AML, the identification of a unifying mechanism is critical to develop therapies with broad clinical efficacy. IL‐1β was recently shown to have the most profound effect on primary AML cell growth in a panel of 94 cytokines 91. IL‐1β responses clustered with those of GM‐CSF and IL‐3, two other AML growth‐inducing cytokines. The elevated responsiveness of IL‐1β‐sensitive AML bone marrow cells appeared to correlate with a significantly increased expression of IL‐1R1 and IL‐1RacP compared to IL‐1β non‐sensitive AML samples, consistent with other findings demonstrating that expression of IL‐1RAcP is a prognostic marker of AML 91, 92, 93. In addition to relaying growth signals from IL‐1 cytokines, a recent study suggested a broader role for IL‐1RAcP in AML by amplifying oncogenic AML pathways through recruitment of the tyrosine kinases FLT3 and c‐KIT in AML cells 92. This suggests that therapeutic neutralization of IL‐1RacP may permit to simultaneously target multiple oncogenic pathways in AML.

Multiple myeloma is a clonal B‐cell neoplasm that results from the growth of malignant plasma cells within the bone marrow 94. Although IL‐1β may be expressed by plasma or bone marrow cells of virtually all patients with MM, it is not produced by normal plasma cells 95, 96, 97, 98. The aberrant IL‐1β produced in MM induces IL‐6 production by bone marrow stromal cells, which in turn supports the growth and survival of the myeloma cells 99. Interestingly, neutralization of IL‐1 signalling in patients with smouldering and indolent myeloma reduced IL‐6 production and prolonged progression‐free survival 100. Indeed, IL‐6 is a central growth factor for myeloma cells, and IL‐1β appears to be a major cytokine that drives paracrine production of IL‐6 by marrow stromal cells in support of MM growth 99, 101, 102, 103, 104.

Role of IL‐18 in cancer

Although IL‐1β and IL‐18 both are pro‐inflammatory members of the IL‐1 cytokine family, their biological functions differ substantially. IL‐18 does not trigger a febrile response, and unlike the restricted expression profile and transcriptional induction of IL‐1β, it is constitutively expressed by most cell types including myeloid cells, fibroblasts and epithelial cells. IL‐1β is virtually absent in the circulation, whereas low picomolar concentrations of IL‐18 are readily detected in the bloodstream of apparently healthy individuals with levels gradually declining with age 52. IL‐18 was originally purified as the IFN‐γ‐inducing factor (IGIF) based on its ability to enhance IFN‐γ production from Th1 cells in the presence of anti‐CD3 antibodies 105. The bioactivity of inflammasome‐matured IL‐18 is under tight control of a high‐affinity binding protein (IL‐18BP) in the extracellular environment. In serum of healthy subjects, IL‐18BP is present at a 100‐fold molar excess over IL‐18, but in several diseases, the concentrations of free IL‐18 increase relative to those of IL‐18BP 53. Notably, high levels of IL‐18 were detected in tumour tissue and in circulation of patients suffering from malignancies, including cancers of the including cancers of the gastrointestinal tract, breast cancer and MM, among others 106, 107, 108. In some instances, high levels of IL‐18 were associated with advanced tumour stage or poor prognosis, suggesting a pro‐tumorigenic role for IL‐18 (Fig 3). However, care should be taken when interpreting IL‐18 levels since some assays show substantial cross‐reactivity between proIL‐18 and the free, bio‐active form of IL‐18. Moreover, they may not discriminate between IL‐18 bound to IL‐18BP and free, bio‐active IL‐18. As a result of the complexity of unambiguous detection of bio‐active IL‐18 and its diverse and context‐dependent biological functions, further investigation of the pro‐tumorigenic and anti‐tumorigenic effects of IL‐18 is warranted 109, 110, 111.

Pro‐tumorigenic role of IL‐18 in solid tumours

Initial studies have assessed the role of IL‐18 in B16 melanoma metastasis by dosing mice with recombinant IL‐18BP. A single intraperitoneal dose of IL‐18BP given 30 min before intrasplenic injection of murine B16 melanoma cells reduced the number of hepatic metastatic foci by 75% and metastatic volume by 80% 112. Moreover, IL‐18BP treatment of mice with established micrometastases resulted in a 25% decrease in metastasis numbers and a 40% decrease in metastasis volume. Tumour‐derived IL‐18 increased the immunosuppressive NK cell fraction and induced PD‐1 expression on these NK cells 113. Similar findings were observed in breast cancer models, and tumour‐derived IL‐18 levels appear to correlate with a negative prognosis in patients with triple‐negative breast cancers 114. Collectively, these findings suggest a significant role for IL‐18 in driving hepatic metastasis and suggest that IL‐18 neutralization may confer anti‐metastatic benefits in patients 65, 71, 72, 112.

The complex role of IL‐18 in colon carcinogenesis: Is it pro‐ or anti‐tumorigenic?

IL‐18 is expressed by lamina propria cells and epithelial cells that line the intestinal mucosa. However, whether IL‐18 confers protection or exerts a pro‐tumorigenic role in colon carcinoma development is unresolved. Consistent with its role in promoting anti‐tumorigenic interferon‐γ production, Th1 cell, NK and cytotoxic T‐cell (CD8⁺ T cells) responses 105, 115, 116, mice lacking IL‐18 or its cognate receptor were reported to be hypersusceptible to colitis‐associated tumorigenesis in the murine azoxymethane (AOM)/dextran sodium sulphate (DSS) model of inflammation‐associated colorectal cancer 117. Consistently, other studies showed that AOM/DSS‐treated NLRP3^−/−, Nlrp3^R258W, ASC^−/− and caspase‐1^−/− mice had an increased tumour burden 118, 119, 120. Conversely, another study 121 failed to observe changes in colorectal cancer development in AOM/DSS‐treated NLRP3^−/− mice, arguing against a tumour‐suppressive role for IL‐18. Thus, future work should delineate the environmental conditions and immune mechanisms that determine how IL‐18 signalling modulates inflammation‐associated colon tumorigenesis.

Pro‐tumorigenic role of IL‐18 in haematological malignancies

Apart from being involved in solid tumours, IL‐18 also seems to play a role in haematological malignancies. A study analysing IL‐18 gene expression in 47 patients with AML suggested that IL‐18 overexpression reflects the convergence of several important unfavourable prognostic factors in AML, including disease status, age and CD34 expression 122. Moreover, IL‐18‐deficient mice were shown to be protected in a mouse model of MM, and IL‐18 was proposed to act by modulating CD8⁺ T‐cell responses and immunosuppressive MDSC generation. This hypothesis was further corroborated by a comprehensive analysis of the transcriptional landscape of the immune microenvironment in MM patients, which revealed a positive correlation between IL‐18 and PMN‐MDSC signature genes and an inverse correlation between the PMN‐MDSC signature and a cytotoxic gene signature 123. Moreover, serum levels of IL‐18 correlated with stage III disease and were associated with increased levels of angiogenic cytokines such as VEGF and increased mortality in a cohort of 65 newly diagnosed MM patients 108.

Therapeutic potential of IL‐1β blockade in cancer

Marketed IL‐1‐neutralizing biologics

Three major IL‐1‐blocking therapeutic agents are currently available on the market (Fig 4A). Anakinra is a recombinant version of the natural IL‐1Ra that modulates IL‐1α‐ and IL‐1β‐driven inflammation by competing for IL‐1R1 binding sites. It was initially approved by the US Food and Drug Administration (FDA) for the treatment of rheumatoid arthritis (RA) in 2001. With the advent of more potent anti‐rheumatic biologics, the drug is nowadays mainly used in the treatment of inflammasomopathies and periodic fever syndromes 85. With a molecular weight of 17 kDa, Anakinra is rather short‐lived (plasma half‐life of 4 hours) necessitating daily injections that are often associated with injection site reactions. However, in most cases these minor adverse events subside after a few weeks. The advantage of the short half‐life is that blood levels fall within hours of treatment discontinuation, which may be advantageous in case of acute infections.

Figure 4. Therapeutic neutralization of IL‐1 and IL‐18 signalling.

Anti‐cytokine therapies are often inspired by natural occurring IL‐1 receptor antagonist or consist of neutralizing antibodies. (A) There are currently 3 approved IL‐1 targeted therapies on the market. Anakinra is the recombinant form of IL‐1 receptor antagonist (IL‐1Ra). By occupying receptor binding sites, Anakinra prevents IL‐1 signalling. Canakinumab is a high‐affinity, fully human monoclonal anti‐IL‐1β antibody designed to bind and neutralize the activity of human IL‐1β. Finally, Rilonacept (also known as IL‐1 Trap) is a fully human dimeric fusion protein that incorporates the extracellular ligand‐binding domains of both the IL‐1 receptor components required for IL‐1 signalling (IL‐1R1 and IL‐1RAcP), linked to the Fc portion of human IgG1. By catching soluble IL‐1, this decoy receptor prevents IL‐1 from signalling through its receptor. (B) While for IL‐1β several approved therapeutics are available, to the best of our knowledge no anti‐IL‐18 therapy has been approved. Currently, two IL‐18 targeted therapeutics are being evaluated in clinical trials: an IL‐18‐neutralizing humanized monoclonal antibody (mAb; GSK1070806) and recombinant human IL‐18BP (Tadekinig alfa). IL‐18BP, IL‐18 binding protein; IL‐1Ra, IL‐1 receptor antagonist; IL‐1RAcP, IL‐1 receptor accessory protein.

IL‐1‐targeted therapeutic agents that boast a longer half‐life and fewer injection site reactions also are available. Canakinumab is a high‐affinity, fully human monoclonal antibody that selectively binds and neutralizes the activity of human IL‐1β. With a molecular weight of about 145 kDa, the mean terminal half‐life of Canakinumab was demonstrated to be 26 days, thus necessitating fewer injections (once a month or fewer) compared to Anakinra 124. It was approved for the first time in 2009 in the United States for the treatment of CAPS. Owing to its long half‐life and selectivity for IL‐1β, Canakinumab is being investigated for several chronic diseases. Notably, the recently completed Canakinumab Anti‐inflammatory Thrombosis Outcome Study (CANTOS) trial (clinicaltrials.gov identifier: NCT01327846) demonstrated the efficacy of IL‐1β neutralization in a large cohort of over 10,000 atherosclerosis patients from 39 countries with increased high‐sensitivity C‐reactive protein levels (hsCRP ≥ 2 mg/l) that were at increased risk for secondary stroke and cardiovascular events 125. The study demonstrated that patients in the Canakinumab group who achieved hsCRP < 2 mg/l had a 25% reduction in major adverse cardiac events, whereas no significant benefit was observed in patients with on‐treatment hsCRP ≥ 2 mg/l.

Rilonacept (also known as IL‐1 Trap) is yet another IL‐1‐neutralizing agent that blocks IL‐1‐induced signalling. Rilonacept is a fully human dimeric fusion protein that incorporates the extracellular ligand‐binding domains of IL‐1R1 and IL‐1RAcP linked to the Fc portion of human IgG1. It received FDA approval for the treatment of CAPS in 2008. With a molecular weight of about 251 kDa, Rilonacept was shown to have a terminal apparent elimination half‐life of about 1 week. In terms of half‐life, Rilonacept is therefore intermediate between Anakinra and Canakinumab, generally requiring a weekly dosing scheme for chronic disease control.

Potential effectiveness of anti‐IL‐1 in cancer treatment

Anti‐IL‐1 therapies have a good safety profile with no major adverse events having been reported in over 100,000 patients with rheumatoid arthritis and autoinflammatory diseases that are/have been treated with Anakinra, Canakinumab or Rilonacept. Apart from injection site reactions that are often associated with biologics, mild infections of the upper airways, abdominal pain and headaches are the prime adverse events observed with anti‐IL‐1 therapies 56, 126, 127, 128. Additionally, neutropenia occurs in approximately 2% of Anakinra‐treated patients and has been reported for Canakinumab as well. However, contrary to anti‐TNF therapies 129, reactivation of latent Mycobacterium tuberculosis is rarely seen in patients receiving IL‐1‐neutralizing biologics. Given this favourable safety profile, anti‐IL‐1 therapies could be explored for additional indications.

One field of interest where IL‐1 signalling is emerging as a promising target is cancer therapy. Recently reported results from the aforementioned CANTOS trial have demonstrated the transformative potential of IL‐1β neutralization in cancer prevention and early interception 125. In this phase III study, the primary endpoint was to test whether Canakinumab reduces the risk of cardiovascular events in patients with a history of myocardial infarction and systemic inflammation independently of lipid‐level lowering 125. As a secondary analysis, CANTOS suggested that neutralizing IL‐1β with Canakinumab reduced lung cancer incidence and lung cancer‐associated mortality 130. The rationale to perform this extra analysis was supported mainly by evidence from animal models and in vitro data that—as discussed above—suggested a potential role for IL‐1β in cancer growth and metastasis 56, 85. By design, all 10,061 patients that were included in the study were free of previously diagnosed cancer (other than basal cell skin carcinoma) at trial entry and were followed up prospectively for incident medical events during 3–5 years. Individuals with increased CRP concentrations show increased risk for several inflammatory cancers, most prominently lung cancer. Furthermore, patients with atherosclerosis commonly smoke, which is another major risk factor for cancer development. The exploratory data suggested a dose‐dependent effect of Canakinumab on the occurrence of fatal and non‐fatal lung cancers in the CANTOS population that by design had a higher‐than‐average risk for the development of lung cancer. The investigators noted a dose‐dependent effect with a relative risk reduction of 67% for lung cancer (HR 0.33 [95% CI: 0.18–0.59]) and 77% for lung cancer mortality (HR 0.23 [95% CI: 0.10–0.54]) observed among patients receiving a 300 mg dose of Canakinumab every 3 months. Furthermore, most benefit was gained by smokers and those who achieved the greatest reduction in CRP and IL‐6. The CANTOS trial was, however, not formally designed as a cancer detection or treatment trial, and these potentially transformative findings will need to be carefully replicated in different settings 130, 131.

Prior to the CANTOS trial, inhibition of IL‐1 signalling in cancer interception has been explored with Anakinra in a phase II clinical trial for smouldering and indolent myeloma (clinicaltrials.gov identifier: NCT00635154) 100. Smouldering or indolent myeloma is a precancerous stage that may develop into MM, a nearly always fatal bone marrow cancer. As discussed above, progression into MM involves IL‐1β signalling, and it was hence hypothesized that inhibiting IL‐1 signalling with Anakinra would prevent or delay active disease. Forty‐seven high‐risk patients with smouldering or indolent myeloma were enrolled in this study and were given Anakinra daily for 6 months. Patients with clinical improvement continued on Anakinra, whereas stable patients and patients at risk of disease progression received low‐dose dexamethasone on top of Anakinra, which was the case for 53% of the patients enrolled in the study. The primary endpoint of the study was progression‐free survival. For the 47 patients who received Anakinra (with or without dexamethasone), a median overall progression‐free disease of 3 years was observed, and in eight patients, this extended to more than 4 years 100. These findings suggest that IL‐1 blockade may significantly delay or prevent progression into active disease. In addition, the results suggested that an on‐therapy reduction in CRP may predict stable disease because patients with a decrease in serum CRP of 15% or more after 6 months of Anakinra monotherapy had a progression‐free survival of > 3 year compared to only 6 months for patients with < 15% drop in serum CRP levels. Notably, a long‐term follow‐up analysis of this cohort showed that median progression‐free survival in patients with a reduced 6‐month vs. baseline CRP level was more than 3 years compared to 1 year for patients with a marginal reduction in CRP 132. These results suggest that IL‐1 blockade in patients with smouldering or indolent myeloma may prevent conversion of latent disease into active MM 56, 85, 100, 132. Considering the increasing incidence of MM in the ageing population, these findings may have substantial impact on the management of this fatal cancer if confirmed in larger placebo‐controlled studies.

Anti‐IL‐1 therapies may also support cancer care by overcoming some of the adverse effects associated with modern immunotherapies. Chimeric antigen receptor‐modified T (CAR T)‐cell therapy represents a novel paradigm for treating cancer 133. Despite impressive cure rates, life‐threatening cytokine‐release syndrome (CRS) and neurotoxicity are frequent adverse effects of CAR T therapies that are clinically associated with high fever, increased levels of acute‐phase proteins and respiratory and cardiovascular insufficiency. Notably, while blocking IL‐6 failed to protect mice from delayed lethal neurotoxicity, recent studies demonstrated that Anakinra abolished both CRS and neurotoxicity in mouse leukaemia models without hindering anti‐leukaemic efficacy and hence substantially extending leukaemia‐free survival rates 134, 135. These studies highlight the potential benefits of combining IL‐1β blockade with cancer immunotherapies to overcome treatment‐induced adverse effects.

Therapeutic potential of IL‐18 neutralization in cancer

Unlike for IL‐1β, no IL‐18‐targeting therapies have been marketed, although two products that neutralize IL‐18 signalling are currently being evaluated in clinical studies (Fig 4B). The IL‐18‐neutralizing humanized monoclonal antibody GSK1070806 was first tested in cohorts of healthy and obese subjects for its safety, tolerability, pharmacokinetics and pharmacodynamic profile 136. Next, a study was designed to investigate the therapeutic benefit of GSK1070806 in the treatment of type 2 diabetes mellitus. While the antibody was well‐tolerated, the compound did not reveal any improvement in glucose control 137. Recombinant human IL‐18BP is another IL‐18‐neutralizing biologic that is currently being evaluated in clinical trials. Phase I studies in healthy volunteers and patients suffering from rheumatoid arthritis and plaque psoriasis showed a favourable safety profile 138. These findings set the stage for a phase II clinical trial that is assessing the potential of recombinant human IL‐18BP in adult‐onset Still's disease and a pivotal phase III clinical trial in patients suffering from potentially life‐threatening inflammation caused by gain‐of‐function mutations in the inflammasome sensor NLRC4 (clinicaltrials.gov identifier: NCT03113760) 139. However, no studies examining the potential use of IL‐18 blockade in cancer patients have been initiated to the best of our knowledge.

Concluding remarks

Although IL‐1β and IL‐18 have been cloned more than 20 years ago, the past decade of inflammasome research has provided remarkable progress in understanding the molecular mechanisms that regulate their production, and an unprecedented level of insight was gained in how they contribute to human diseases. As reviewed here, a wealth of preclinical data has been generated over the years suggesting both pro‐ and anti‐tumorigenic roles of IL‐1β and IL‐18. IL‐1β and IL‐18 have emerged as key inflammatory cytokines that contribute to rheumatic, cardiovascular and metabolic diseases. Among other studies, the CANTOS trial further highlighted a potentially transformational effect of IL‐1β neutralization in lung cancer development. These promising results are likely to propel further investigation of the clinical benefit of IL‐1β‐ and IL‐18‐neutralizing agents in a host of malignancies and other diseases (see also Box 1). Moreover, fast‐paced research in the inflammasome field is likely to provide additional targets that can be drugged for therapeutic modulation of IL‐1β and IL‐18 production. Specifically, the discovery of small molecule inhibitors that target the NLRP3 inflammasome suggests that inflammasome biology may be exploited to target production of IL‐1β and IL‐18 by oral therapies 140, 141. These endeavours should ultimately provide additional treatment options for patients suffering from inflammatory diseases and malignancies alike.

Conflict of interest

M.L. is an employee of Janssen Pharmaceutica. The authors declare that they have no conflict of interest.

Box 1: In need of answers.

1. How do systemic levels of IL‐1 and IL‐18 impact on tumour growth and metastasis, and how do they impact anti‐tumour immunity and the local inflammatory milieu in the tumour microenvironment?

2. What is the role and relative contribution of tumour, stromal and infiltrating hematopoietic cells to IL‐1β and IL‐18 secretion in the tumour microenvironment along different stages of tumorigenesis and cancer development? Analysis of cell type‐specific and inducible deletion of IL‐1β, IL‐18 and their cognate receptors in mouse models of cancer may represent attractive approaches to address this question.

3. What are the upstream mechanisms that drive maturation and secretion of IL‐1β and IL‐18 in tumours, and which environmental factors modulate their activation? The analysis of inflammasome activation in primary human cancers, syngeneic animal models with an intact immune system and genetically engineered mouse models of cancer should provide important insights into inflammasome signalling during the oncogenic process.

4. What are the dominant pro‐ and anti‐tumorigenic mechanisms by which IL‐1β and IL‐18 regulate different stages of cancer development?

5. How important is pyroptosis in the release of IL‐1β and IL‐18 in the tumour microenvironment and during different stages of cancer development?

6. In which patients can inflammasome activation and IL‐1β/IL‐18 signalling be targeted most effectively for cancer prevention and treatment?

EMBO Reports (2019) 20: e47575