Show Summary Details
Page of

Cancer immunology 

Cancer immunology
Chapter:
Cancer immunology
Author(s):

Herman Waldmann

DOI:
10.1093/med/9780198779452.003.0023
Page of

PRINTED FROM OXFORD MEDICINE ONLINE (www.oxfordmedicine.com). © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 20 September 2019

Why cancer immunology?

The reader may wonder why cancer immunology should be considered a topic distinct from immunology in general, or from transplantation immunology from which many fundamental concepts of the immune system have evolved. In confronting pathogens or transplants the immune system is often rapidly exposed to large doses of non-self-antigens in the context of molecular adjuvant-creating moieties (so-called molecular patterns) that elicit multiple cascades of immunity (Matzinger, 1994; Medzhitov and Janeway, 1997). In contrast, it is not easy to imagine how a ‘self’ cell which has recently developed sufficient mutations to become malignant, can provide adequate adjuvant-creating signals, let alone non-self-antigens, to evoke equivalent destructive immune responses. As the malignancy expands what cues, if any, can it give the immune system that it should respond?

Early conjectures were that growing ‘self’ tumour cells could somehow ‘sneak through’ the various pro-immunity checkpoints, so bypassing immune reactivity.

The issue of tumour immunogenicity has been, and continues to be, a dilemma for immunologists. Leaving aside virally induced tumours, would we not be tolerant of the protein’s tumour expressed in our own tumours? Even if we were not naturally tolerant of some tumour antigens, would these not be shared with normal tissues? Presumably then, tumour immunity might also risk autoimmunity?

Although answers to these questions are still incomplete, the explosion in fundamental immunology has, over the past 25 years or so, provided the groundwork for a better understanding of the relationships between tumours and the immune system. The advent of monoclonal antibodies, both as diagnostics and as probes for molecular function, have been important. The identification of dendritic cells as a major intermediary between the antigen source and T-cell activation has been crucial (Banchereau and Steinman, 1998). Major advances in molecular biology and the creation of mutant mice lacking defined gene products, has pinpointed key molecules influencing immune function. Finally, many translational efforts in vaccination, autoimmune disease, and transplantation have enabled identification of hitherto undervalued mechanisms that the immune system uses to regulate itself. Historically, these had been poorly studied because it was easier to quantitate immune responses than the lack of them (i.e. self-tolerance). Moreover, as self-tolerance was largely seen as a fait accompli from birth, research into mechanisms of self-tolerance has always been particularly demanding.

Against the backdrop of advances in fundamental immunology, there have been several clues that the immune system could attack cancers, promoting their rejection as if they were ‘foreign’ transplants. From all these contributions, the notion that tumours simply sneak through the immune system has changed. We now accept that some tumours can evoke and/or provide sufficiently adjuvant-like signals which, if adequately sensed, can promote immune reactivity in the host. With this accepted a new hallmark of cancer can be defined as the cross-talk between tumours, their supportive stroma, and the immune system, that impedes immune reactivity using diverse inhibitory mechanisms acting throughout the evolution of that tumour (Hanahan and Weinberg, 2011). Such suppressive mechanisms, formerly studied solely in relation to carcinogenesis, are now becoming increasingly relevant to immunoregulation more generally (Zitvogel et al., 2013; Belvin and Mellman, 2015). At a practical level, an improved understanding of these mechanisms has suddenly created a major growth industry in practical efforts to give the immune system the upper hand. These range from conventional small molecule drugs, antibodies, cytokines, and cellular therapies, as well as a recognition that what were formerly thought of as chemotherapeutic drugs, are, in retrospect, agents that can also liberate host antitumour immune function in diverse ways. In short, our concepts of tumour immunology and our application of them are developing fast, being both informed by, but also informing, fundamental immunology.

The immune system and protection against pathogens

The vertebrate immune system has evolved to protect against microbial infection while minimizing damage to self. To achieve this, it uses innate mechanisms mediated by plasma proteins and diverse cells to contain infection long enough for the adaptive system of antigen-specific lymphocytes to engage, and to generate responses, best able to clear the microbe. The innate mechanisms have no specificity for the foreign antigen, but reflect recognition by preformed receptors of one or more forms of perturbation, including microbially-encoded molecular patterns, so-called pathogen-associated molecular patterns or PAMPS (Medzhitov and Janeway, 1997; Beutler and Rehli, 2002), comprising proteins, lipids, carbohydrates, and/or nucleic acids, and self-encoded molecular beacons of ‘cell stress’ or danger. The consequent innate effector mechanisms include activation of complement, enhancement of phagocytosis, and exocytosis by neutrophils, eosinophils, basophils, macrophages, natural killer (NK), γδ‎ T cells, and other cells, all aimed at elimination of the pathogen.

In contrast to the rapidly-activated, generic innate response, the adaptive response is highly-specific but delayed. It depends upon the expansion and differentiation of clones of (thymic-derived) T cells and (bursa-equivalent) B cells that are selected by the binding of antigen to clonally distributed receptors produced by seemingly-random somatic gene rearrangement within T-cell and B-cell progenitors (Kurosawa and Tonegawa, 1982). Following their antigen-driven selection, the activated B cells and T cells generate differentiated progeny with effector function, and additionally commit cells to reservoirs that retain memory for the eliciting antigens. Whereas T-cell receptors of all progeny stay true to the parental cell, B-cell (immunoglobulin) receptors undergo somatic mutation of their heavy and light variable (V-) regions to derive better fitting (higher affinity) receptors, culminating in best-fit antibody molecules being released into the circulation. Antibodies of identical specificities can be made using distinct Heavy (H-) chains (classes and subclasses or isotypes) responsible for mediating different functions, some of these being critical to elimination of the organism or neutralization of its key attachment and virulence molecules, through localizing and amplifying innate mechanisms. Conversely, T-cell receptor structures are constant, but can be expressed by T cells of different effector potentials, thereby suiting the host response to the nature of the challenge, as is elaborated on next.

To a first approximation, the adaptive system uses antibodies to target pathogens that exist outside the body’s cells but exploits T cells to deal with microbes that reside within cells. In most vertebrates studied in any detail, the great majority of circulating T-cells bear receptors (TCR) made up of so-called α‎ and β‎ chains, although this is not the case in, for example, chickens and in cattle during the first year of life, in which very many T cells bear TCRs composed of γ‎ and δ‎ chains. Moreover, seemingly in all vertebrates, T cells that reside within non-lymphoid tissues (e.g. skin and gut), are enriched in γδ‎ T cells distinct from those in the circulation.

For αβ‎ T cells to recognize intracellular protein antigens, it is necessary for the antigens to be processed into shorter peptide fragments which are loaded intracellularly into the clefts of major histocompatibility complex (MHC) Class I or MHC Class II molecules, thereby stabilizing those MHC molecules and their expression at the cell surface. MHC Class I-associated peptides provide the target for CD8 + T-cell-mediated cytotoxicity, while MHC Class II-associated peptides provide the target for CD4 + T cells.

In order for T cells to be activated, it is essential that they see their peptide antigen on activated ‘antigen-presenting cells’, particularly dendritic cells (DC) (Banchereau and Steinman, 1998; Mellman, 2013), which are specialized forms of innate myelomonocytic cells. Initial and necessary activation or ‘licensing’ of DC largely occurs through the previously mentioned innate-sensing receptors. Among these, Toll-like receptors (TLRs) are the best studied (Medzhitov and Janeway, 1997; Beutler and Rehli, 2002). Suffice it to say, the sensing of pathogens by engagement of TLRs and/or other equivalent receptors is crucial to the licensing of DC that is, in turn, critical for their capacity to activate naïve T cells. Licensing is manifest in an altered DC phenotype involving, in particular, upregulation of sets of costimulatory molecules such as CD40, CD80, and CD86. These interact with corresponding receptors on T cells to provide so-called ‘second signals’ that enable antigen to drive T cell activation. In short, it is the second signal that connects innate immunity to adaptive immunity. Without such licensing DC, antigen presentation will not only fail to immunize T cells but may even end up tolerating them. In fact, DC can be ‘decommissioned’ away from this licensing function in many ways, including exposure to diverse agents such as IL-10, TGFβ‎, Vitamin D, and interaction with regulatory T cells (Steinman et al., 2003; Hammer and Ma, 2013).

As was introduced earlier, antigen-specific CD4+ T cells can differentiate into subsets with different immune functions. Th1 cells making IFNγ‎, Th2 cells making IL-4, Th17-cells making IL-17, and T cells expressing Foxp3 and able to regulate other immune cells. These different functional subsets are responsible for orchestrating and sculpting the behaviour of diverse cellular elements through their interaction with MHC Class II-expressing myelomonocytic cells of which DC are key players. Not only can DC be licensed by diverse sensors of perturbation, but they can also be activated by CD4+ T cells, as a route to providing help to enable further cohorts of naïve CD4+ or CD8+ T cells to be recruited into immune responses (Ridge et al., 1998). Among these, CD8 + T cells are best known for their cytotoxic functions, but they too can secrete influential cytokines affecting their local milieu.

For an adaptive immune response to occur, antigens must gain entry to DC undergoing activation-induced enhancement of antigen processing and presentation. To facilitate their interactions with (their so-called ‘priming of’) naïve T cells, licensed DC migrate from the tissues, within which they were exposed to pathogens, moving to draining lymphoid tissues, wherein they promote the clonal proliferation and differentiation of those T cells with TCRs specific for the antigens presented by the licensed DC Thereafter, primed T cells re-enter the circulation, with some making their way to the peripheral sites where the initiating antigens are located. The clues that direct them to the correct locations arise from the inflammation evoked by the infection. Local inflammation induces the neighbouring post-capillary venules to upregulate distinct patterns of adhesion molecules (as if barcodes) which can divert the recently activated T cells and other leukocytes out of the circulation at that site and no other. Leukocytes with different functions perceive different barcodes, so determining the functional composition of white cells entering the tissue. The nature of the infective agent is indeed influential in determining what barcodes are exhibited on the endothelium. Once in the tissues, both CD4+ and CD8+ T cells can liberate influential cytokines, interact with tissue-resident and other tissue-infiltrating immune cells, and deliver protective functions. Altruistic self-sacrifice of virally infected tissue cells that can be easily replenished provides an effective way to eliminate pathogens.

While a proportion of T cells are recruited to peripheral tissues, others may be retained in the lymphoid tissues, as in the case of some CD4 helper T cells that localize to germinal centres where they provide cytokines to help antigen-specific B cells generate antibodies. Other CD4 + and CD8 + T cells may differentiate as memory cells, using their own particular barcodes to locate to sites appropriate for their role.

These steps to achieve beneficial pathogen-directed immunity can be considered as necessary stages for virtually all forms of adaptive immunity associated with αβ‎ T cells.

As might be expected, the immune system has inbuilt checks and balances to prevent inflammation-generating immunity getting out of hand. In part, this is achieved by a wide range of co-inhibitory molecules that counter secondary signals induced by licensed DC (Chambers and Allison, 1999). The best known of these are CTLA4 and PDL1, molecules which have been extensively studied in their capacity as ‘checkpoint inhibitors’. Understanding the circumstances where co-inhibition dominates costimulation offers a major theme in cancer immunology.

Whereas most αβ‎ T cells provide the means to monitor specific intracellular microbes by their response to processed peptide antigens presented by MHC, other αβ‎ T cells can sense microbial lipids presented by a set of MHC-related molecules, known collectively as CD1. Moreover, the repertoire of moieties sensed by T cells is increased further by γδ‎ T cells whose recognition is not limited to either MHC or CD1-associated antigens. Furthermore, activation of γδ‎ T cells does not necessarily require elaborate antigen processing, and presentation, and consequent clonal outgrowths. Consequently, γδ‎ T cells are not limited to the delayed adaptive response mode of αβ‎ T cells. Indeed γδ‎ T-cell recognition of self-encoded beacons of cell and tissue dysregulation (Willcox et al., 2012) can provide a rapid mechanism for the immune system to eliminate stressed cells in tissues, such as those infected by pathogens, or those subject to carcinogens (Willcox et al., 2012; Vantourout and Hayday, 2013). Thus, γδ‎ T cells illustrate how some lymphocytes can contribute both to the early innate phase and to the delayed adaptive phase of an immune response. The same may be true of some subsets of B cells. Importantly, we currently lack an understanding of how the responses of such innate-like lymphocytes resident within tissues impacts upon the temporal progression of the immune response through its innate and adaptive phases, and the development of antigen-specific memory.

Relevance of antipathogen immunity to cancer immunology

Given the necessary stages required for T cells to become activated, to deliver effector functions, and to generate challenge-specific memory, what would be needed for tumour antigens to achieve the same outcome?

The necessary elements would be:

  1. 1. antigens to which T cells with cognate TCRs are available in the circulating immune repertoire;

  2. 2. appropriate molecular signals of cell/tissue perturbation to license dendritic cells;

  3. 3. absence of tumour-associated elements that might interfere with licensing, or even decommission DC;

  4. 4. adequate bar-coding of primed T cells to permit their homing to sites of tumour growth and to sites of memory reservoirs;

  5. 5. receptiveness of the tumour and its vasculature to entry of primed T cells;

  6. 6. a tumour-associated environment (TAM) that enables the effector T cells to express their full effector potentials.

Failures in routinely delivering destructive immunity to cancers suggests constraints at one or more of these levels. A major theme of modern cancer is to identify routes to bypass any, or all, of these constraints.

Self-tolerance and its relationship to cancer

The immune system adopts three general strategies to minimize reactivity to ‘self’

The first involves tolerising self-reactive lymphocytes as they develop in the primary lymphoid organs. For T cells this happens in the thymus, and for B cells largely in the bone marrow. Remarkably the thymic medullary epithelial cells, through the function of the Autoimmune Regulator transcription factor AIRE, promiscuously express a vast array of self-proteins and their processed peptides, enabling a massive purging of the antiself T cells that engage them at this staging post (Anderson and Su, 2016; Richards et al., 2016).

Simultaneously, the actions of another transcription factor (Foxp3) expressed in some CD4 + T cells can rescue T cells with antiself reactivity while endowing them with a protective regulatory function (Hori et al., 2003; Ramsdell and Ziegler, 2014). These natural or thymic regulatory T cells (tTreg) seem critical for self-tolerance, as their absence in humans and mice can result in severe multisystem autoimmune disease. Clearly, promiscuous gene expression through AIRE does not ensure sufficient clonal deletion to protect against all autoimmunity. This suggests that any residual self-reactive T cells spared from clonal deletion are likely held in check by the action of Treg.

The antigen-dose-sensitivity of B cells to clonal deletion appears less than that of T cells (Chiller et al., 1971; Goodnow and Basten, 1989). This means that B cells against abundant, ubiquitous antigens are deleted or inactivated during development, whereas those against sparse self-antigens may be spared. As B cells need T-cell help to generate antibody responses (Mitchell and Miller, 1968), and as any potential antiself helper cells are normally deleted, T-cell-independent B-cell autoimmunity is unlikely. Whenever IgG autoantibodies arise, there must always have been an antecedent CD4 + helper T-cell response to an antigen (Fig. 23.1)

Fig. 23.1 T cells become tolerant to most self-antigens during their development in the thymus. Expression of the AIRE gene in thymic medullary epithelial cells results in promiscuous expression of the majority of self-proteins with each medullary cell expressing a small proportion of such antigens, while the whole population provide coverage for all. Whether these antigens are expressed on the epithelium or reprocessed by dendritic cells, the outcome is the deletion of most self-reactive T cells. Some self-MHC restricted CD4 T cells, which come to express the transcription factor FoxP3, are not deleted, and emerge from the thymus as regulatory T cells, that provide an additional failsafe against autoimmune disease.

Fig. 23.1 T cells become tolerant to most self-antigens during their development in the thymus. Expression of the AIRE gene in thymic medullary epithelial cells results in promiscuous expression of the majority of self-proteins with each medullary cell expressing a small proportion of such antigens, while the whole population provide coverage for all. Whether these antigens are expressed on the epithelium or reprocessed by dendritic cells, the outcome is the deletion of most self-reactive T cells. Some self-MHC restricted CD4 T cells, which come to express the transcription factor FoxP3, are not deleted, and emerge from the thymus as regulatory T cells, that provide an additional failsafe against autoimmune disease.

The second category of tolerance processes, known as ‘peripheral tolerance’ embraces several mechanisms (Mueller, 2010). For example, a T cell seeing its antigen in the absence of other triggering signals may be inactivated or die, or even become outnumbered by lymphocytes that regulate immune function. The best studied regulatory lymphocytes are CD4 + Treg, some of which (tTreg) arise in the thymus, as previously mentioned, while other Treg are induced de-novo in tissue microenvironments rich in TGFβ‎ and/or lacking pro inflammatory mediators, or in circumstances of mTOR inhibition (Chen and Konkel, 2010; Waldmann et al., 2016). Intimately involved in dictating these pathways are diverse tissue-infiltrating myelomonocytic cells-including monocytes, granulocytes, and the heterogenous population of myeloid suppressor cells (MSDC; see Gabrilovich et al., 2012; Marvel and Gabrilovich, 2015); tumour-induced endothelial cells and other stromal elements (Quail and Joyce, 2013; Joyce and Fearon, 2015); and innate-like tissue-resident lymphocytes (Hayday and Tigelaar, 2003).

Third, and rarely discussed, are the range of ‘silent’ events in normal or healing tissues that discourage or curtail immune reactivity. It is quite likely, given the different physiological demands on different tissues, that each might use different mechanisms to ensure that immune responses to eliminate microbes do not get out of hand and cause irreparable tissue damage. At the extreme end of these are tissues, identified from transplantation research, wherein immune responses are vigorously constrained. These have been referred to ‘immunologically privileged sites’, of which the placenta and anterior chamber of the eye, and to a lesser extent, brain, testis, and ovaries are the best known (Stein-Streilein and Streilein, 2002; Simpson, 2006). Almost certainly all tissues will, through a coordinated activity of their diverse cellular components, create molecular environments that limit immune activity within them. Equally tissues in the process of healing and remodelling provide many negative feedback signals, of which local active TGFβ‎ inducing immunosuppressive adenosine-generating ectonucleotidases on many tissue cells provides one good example (Regateiro et al., 2011).

Tolerogenic processes can also be promoted in circumstances where some immune activation has already occurred. These have been highlighted in the analysis of experimental animals, where foreign tissues have been accepted long term (Baas et al., 2016; Waldmann et al., 2016), and in reversal of organ-specific autoimmune diseases, after short term therapeutic intervention (Chatenoud et al., 1997). Studies of the tissue components in these situations have emphasized a role for newly established microenvironments within such tissues, whereby cross talk between innate, adaptive immune cells as well as tissue cells renders the tissue less permissive for immune attack.

Collectively, these examples of the tissue ‘fighting back’ can be considered as forms of acquired immunological privilege.

Research into mechanisms of natural and acquired privilege provides information on the diversity of protective tissue reactions that can operate. On the one hand, these may reflect the directions any given tissue can follow to sculpt the immune infiltrates within it to ensure damage limitation. On the other hand, they may give clues on the ingredients that tumours use to achieve the same outcome. In short, the study of immune interactions with tumours has lots to learn from the study of immune interactions with normal tissues, and vice versa.

In fact, the analysis of how the tumour microenvironment allows tumours to escape T-cell damage is now offering the most rapidly advancing evidence for privilege mechanisms in general. Undoubtedly, its understanding will have broad ramifications, not only in the treatment of cancer, but also in reversal of autoimmune diseases, the prevention of allogeneic cell and organ rejection, and the limitation of major inflammatory diseases.

The combined knowledge emerging from all these directions, and especially from cancer studies, challenges the view of immunity and tolerance as binary discrete processes. Elements of each are always contributing to the way that the immune system reacts to antigen encounter. What we call an immune response, is one that emerges despite many constraining influences, whereas when we speak of tolerance to tumours, we are recognizing that tolerance mechanisms have simply prevailed over immune ones that were present but that passed unnoticed.

Cells hitherto thought of as mediators of immunity, can, in particular contexts, also behave as mediators of tolerance and vice versa (Wang et al., 2010). The mere enumeration of cell types in biopsy cannot tell us whether the immune infiltrate is an indicator of imminent destruction or of protection. More detailed parameters of cell subpopulations, their relative proportions, their micro-anatomical locations, frequencies of discrete cell-cell interactions, and indicators of their functional capacity, are all relevant in what has come to be called the analysis of cell ‘contexture’ as a guide to cancer prognosis and treatment (Fridman et al., 2012; Galon et al., 2013; Gentles et al., 2015; Galon et al., 2016).

The acceptance of the notion, that most of our well-studied cells of the innate and adaptive immune system, can be a force for tumour rejection as well as for tumour acceptance, is key to therapeutic intervention in a way appropriate to the patient’s tumour. Despite, the possible diversity of tolerance-related mechanisms operating we need, to establish priority lists of the most generic and tractable of these that can be targeted, rather than aiming at treatments that become too personalized. Ultimately, our chosen drugs will need to offer a commercial incentive to their manufacturer, and fiduciary practicality to healthcare providers.

Why do we think that tumours may have antigens the immune system can target?

Evidence for immunosurveillance of cancer

It has long been known that some cancer patients undergo spontaneous remissions of their disease. Patients receiving ‘immunosuppressive’ drugs to prevent graft rejection are more susceptible to diverse cancers. Some cancers are clearly caused by oncogenic viruses (e.g. human papilloma virus (HPV), EBV, or hepatitis), but others seem not to be. The possibility that immune mechanisms might have been involved in control of cancer was recognized some 60 years ago, when it was shown that chemically induced tumours could be rejected in inbred mouse strains. These and related experimental findings spawned the notion of cancer immunosurveillance proposed by Burnet (Dunn et al., 2004). It was initially somewhat discouraging that T-cell deficient mice did not show a higher susceptibility to chemically–induced cancers compared to their normal counterparts, although their increased susceptibility to tumours caused by polyoma virus was consistent with immunosurveillance (Simpson and Nehlsen, 1971). However, later evidence that neutralization of IFNγ‎ by antibodies enhanced tumour growth, and that IFNγ‎ deficient mice were more susceptible to tumours, restored optimism to the concept of immunosurveillance, further boosted by clear evidence that mice deficient specifically in γδ‎ T cells have heightened sensitivity to environmental carcinogens (Girardi et al., 2001). Likewise, tumour incidence is higher in mice in which elements of the T-cell killing machinery and members of the TNF family were compromised.

Later work showed that tumours developing in immunocompromised mice demonstrated greater immunogenicity when transferred into fresh immunocompetent hosts, than those developing in controls. This observation led to the immunoediting hypothesis (Dunn et al., 2004), a modification of the original immunosurveillance theory. Here it was envisaged that the interaction of the immune system with cancers operates in three phases: elimination, equilibrium, and escape. Although no experiment has definitely established that this hypothesis satisfactorily reflects a general facet of cancer biology, it provides a useful framework for considering how the immune system may interact with tumours, and hence how the balance may be swung toward tumour eradication and patient benefit. ‘Elimination’ reflects the capacity of innate and adaptive mechanisms to clear early tumour cells. Should this fail, then the ‘equilibrium’ phase reflects the capacity of T cells to keep the tumour under control. When that eventually fails for whatever reason, tumours enter the ‘escape’ phase. The proposed equilibrium phase offers a period during which the immune system exerts pressure on tumours to develop escape mutants. Many tumours at time of presentation are likely to be in this proposed escape period. Identifying the reasons for tumour escape may offer therapeutic routes to its prevention.

As mentioned, evidence supporting the immunoediting idea has been provided from chemically induced and spontaneous tumours in mice, based on identifying changes in immunogenic antigens. In recent years, abundant data from human tumours has been consistent with a role for the immune system in shaping tumour evolution. Analysis of the quality, location, and intensity of immune infiltrates in tumours has provided prognostic information for tumour outcomes with a series of Th1 related indicators proving a more favourable prognosis for many tumours. Surprisingly, ‘signatures’ reflecting Th2-like, Th17, or regulatory T-cell signatures have not been so clear cut. Perhaps consistent with the mouse data mentioned earlier, a γδ‎ T-cell signature has proved to be the strongest correlate of overall survival in an unprecedentedly large survey of 18,000 patients (Gentles et al., 2015). Understandably, many attempts are currently being made to take many of the immune parameters into account in providing an ‘immunoscore’ to predict the potential tumour outcome (Galon et al., 2016; Mlecnik et al., 2016).

The nature of tumour antigens

Early evidence for immunogenic tumour-associated antigens came from serology of individuals bearing tumours, and in paraneoplastic immune disorders with antitumour antibodies detecting shared antigens on neuronal cells. Although indicative of adaptive immunity, little evidence existed on whether the response was driven by self-antigens, or neoantigens, derived from altered self-proteins encoded by genes mutated by genotoxic carcinogens. Spontaneous antitumour T-cell responses (both CD4+ and CD8+ T cells) have been identified in some patients with melanoma, and as mentioned earlier, these examples give indications that the host may be able to mount response to self-antigens in tumours, where perhaps clonal deletion had not occurred.

Virally induced tumours (e.g. EBV, HPV, hepatitis virus) all offer potential tumour-associated antigens that are ‘foreign’, to which the host will not be naturally tolerant. The natural CD8 T-cell mediated immune-surveillance that prohibits malignant B-cell accumulation in most EBV infected people is a good indicator that antiviral immunity enables a long-term equilibrium between host and cells harbouring the oncogenic virus.

However, the most compelling evidence that neoantigens emerge in cancers, that these are subject to immune attack, and that they change over the time course of tumour development, comes from advances using modern DNA recombinant technologies involving exome sequencing of tumours and T cells infiltrating them, in-silico extrapolations of whether peptides predicted to derive from mutated proteins are potential immunogens able to bind to MHC molecules, and technologies to show that T cells can bind these and become activated (Linnemann et al., 2014; Gubin et al., 2015; McGranahan et al., 2016; Schumacher and Schreiber, 2015; Ward et al., 2016). After successful studies in mouse cancer models, such technology is now being applied to human cancers, albeit with a propensity of studies largely, thus far, confined to melanoma (Fig. 23.2).

Fig. 23.2 Identification of cancer neoantigens. Ribonucleic acid (RNA) sequencing of tumour extracts RNA sequencing data enables detection of mutations in expressed genes. Likely mutated peptides are catalogued, and those likely to bind to MHC Class I are predicted and tested for possible immunogenicity in in vitro functional assays. Similar approaches are used to detect MHC Class II binding neoantigens that stimulate CD4 + T cells.

Fig. 23.2 Identification of cancer neoantigens. Ribonucleic acid (RNA) sequencing of tumour extracts RNA sequencing data enables detection of mutations in expressed genes. Likely mutated peptides are catalogued, and those likely to bind to MHC Class I are predicted and tested for possible immunogenicity in in vitro functional assays. Similar approaches are used to detect MHC Class II binding neoantigens that stimulate CD4 + T cells.

Source: data from Schumacher TN and Schreiber RD, ‘Neoantigens in cancer immunotherapy’, Science, Volume 348, Issue 6230, pp. 69–74, Copyright © 2015 American Association for the Advancement of Science.

Many neoantigens have now been identified that bind to both MHC Class I and II molecules. It is also clear that some tumours exhibit a high mutational load, and others less so. The expectation is that those with a higher mutational load will generate more neoantigens. There is now the opportunity to pool such information with that of the contexture scores, as well as therapeutic outcomes to assess the impact of neoantigenic load on outcomes of therapy (Rizvi et al., 2015). As expected, given the stochastic nature of mutational neoantigens, it may be difficult to find broadly-shared antigens within tumour types. That, together with MHC polymorphisms may make it difficult to identify worthwhile tumour-vaccine candidates, other than on a personalized basis. In the context of personalized therapies (if such were commercially viable), knowledge of an individual’s neoantigen repertoire may enable selection and expansion of antigen-specific cells from those infiltrating the tumour, or engineering of T cells to express TCRs recognizing such antigens for adoptive immunotherapy.

A word of caution though! These aspirations may be confounded by limitations imposed by the patient’s available TCR repertoire, as well as our lack of knowledge of how both qualitative and quantitative MHC/peptide epitope expression at the cell surface determines immunogenicity.

The demonstration of tumour-specific T cells in tumours

The strategy for studying T cells infiltrating tumours owes much to early studies of adoptive T-cell transferred from mice immunized to tumour antigens. Poor protection was initially observed until transfers were attempted into lymphocyte-depleted recipients. Pioneering studies by Rosenberg using extracted human melanoma infiltrating lymphocytes expanded in culture with IL-2, and transferred into lymphocyte-depleted patients provided evidence that these cells could attack the cancer and persist in the circulation (Rosenberg and Restifo, 2015; Topalian et al., 2015). The necessity for prior host lymphocyte depletion for efficacy could reflect a need for homeostatic cytokines and ‘space’ to enable transfused lymphocytes to expand further and differentiate, or relate to, ablation of other immune inhibitory pathways operating in the lymphocyte-replete host.

These studies with TILs have confirmed that melanomas contain T cells that recognize both ‘self’ as well as mutated antigens. Studies with epithelial cancers, using exomic sequencing to identify possible neoantigens, have provide evidence that TILs include T cells responsive to some of these neoantigens. Once expanded in vitro, some such TILs could achieve dramatic tumour regression on transfer (Hinrichs and Rosenberg, 2014; Rosenberg and Restifo, 2015). Another approach to identify reactive T cells has been to engineer their receptors (TCR) so that they can be transfected into naïve T cells to give them target specificity (Rosenberg and Restifo, 2015). In some examples using TCRs to non-mutated antigens, where cancer regressions were observed, damage to other normal tissues could also be seen, indicating broader display of those antigens beyond the tumour itself. Somehow, the host from which such TILs had been derived, had not exhibited such off-target damage-perhaps pointing to protective regulatory processes operating to limit damage. Nonetheless, it is the case that some melanoma patients with immune responses to their tumours present clinically with vitiligo in normal areas of skin.

How do tumours provide the innate signalling mechanisms in the way that microbial pathogens do?

It is now widely acknowledged that there exist a large range of mechanisms for sensing cell stress, damage, and death, which can provide adjuvanticity for tumour antigens, in the way that PAMPS do for microbial antigens. These damage-associated molecular patterns (DAMPs) are distinct entities but can engage in much cross talk with PAMP signalling mechanisms. A detailed understanding of DAMPS and their signalling pathways will be an important avenue of future research aimed at maximizing the immune response to tumour antigens (Schaefer, 2014; Woo et al., 2015).

The tumour-associated microenvironment limits immune activity against tumour antigens

The accumulating evidence, that some tumours contain T cells with specificity to tumour neoantigens, raises the question of why these are not eradicating the tumour. A compelling answer to this question has recently been provided by the massive and effective immune response to some tumours unleashed by blockade of known co-inhibitory molecules such as PD1 or CTLA4 expressed on T cells. These findings, together with previously mentioned prognostic studies on the nature of tumour infiltrates, have clearly indicated that the tumour-associated microenvironment, shaped both by the tumour and by immune components, determines the extent to which T cells can exert their antitumour potential. As each contributor is defined, then additional novel targets for checkpoint blockade become available.

Cells infiltrating tumours may have the potential to contribute to tumour damage, but this may be outweighed by other contributions to contribute tumour growth, and to immune privilege (Hanahan and Coussens, 2012; Hagerling et al., 2015; Joyce and Fearon, 2015; Marvel and Gabrilovich, 2015; Quail and Joyce, 2013). Growth promotion may be mediated by diverse growth factors, and proteolytic enzymes that might alter the extracellular matrix may permit tumour cell spread. Cell sources contributing such influence are not limited to leukocytes but may also include tumour-associated fibroblasts and blood vessels, for example.

As indicated earlier, T-cell priming requires licensing and migration of antigen-presenting cell (APC). There is much evidence to indicate that DC isolated from human tumours are functionally impaired, if not decommissioned. Various products in the TAM have been implicated, including cytokines such as TGFβ‎, IL-10, and vascular endothelial growth factor (VEGF), macrophage colony-stimulating factor (MCSF), and even hypoxia, in some cases associated with enhanced expression of the co-inhibitory molecule PDL1.

Even if T cells have been primed, they need to gain access to tumour cells. This requires that they gain entry through the tumour vasculature and migrate along chemotactic gradients to their targets. Failure of tumour vasculature endothelial cells to properly coordinate expression of key adhesion molecules is one element to tumour protection. VEGF, endothelins, and nitric oxide have been implicated as key mediators at this checkpoint, in addition to a number of well established immunosuppressive molecules.

Beyond the endothelium there are many other cell types that can prevent immune engagement, including Treg, myeloid-derived suppressor cells (Gabrilovich et al., 2012; Marvel and Gabrilovich, 2015), and monocytes that are ‘alternatively activated’ as if exposed to Th2 T-cell products (Hanahan and Coussens, 2012). Some of these effects on effector T cells can be mediated by enzymes catabolizing among others, tryptophan (Munn and Mellor), and L-arginine (Molon et al., 2011) creating a nutrient-deprived local microenvironment, and generating range of anti-inflammatory moieties, such as kynurenines, active radicals of oxygen and nitric oxide, and peroxynitrite. Peroxynitration and inactivation (by nitration/nitrosylation) of chemokines and cancer neoantigens is one proposed mechanism by which products of arginine metabolism can impede immune functions. All these factors influence the balance of cell types that accumulate in tumours, and the quality of immune privilege that results.

Analysis of the components of the tumour microenvironments indicate widespread cross talk between cancer cells, their stroma, and the immune system (Fig. 23.3). Evidence is accumulating that diverse signalling pathways influencing tissue development and carcinogenesis (e.g. Wnt; see Yamabuki et al., 2007; Sato et al., 2010), as well as tumour suppression (e.g. p53, PTEN, RB1, and ARF; Munoz-Fontela et al., 2016), have profound local immune subversive functions in addition to their cell intrinsic functions.

Fig. 23.3 (A–F) Mechanisms of immune suppression within the tumour microenvironment. Diverse cellular components in the tumour microenvironment regulate the entry, accumulation, activation, and expansion of T cells in tumours exemplified in the representative frames shown.

Fig. 23.3 (A–F) Mechanisms of immune suppression within the tumour microenvironment. Diverse cellular components in the tumour microenvironment regulate the entry, accumulation, activation, and expansion of T cells in tumours exemplified in the representative frames shown.

Reproduced with permission from Joyce JA and Fearon DT, ‘T cell exclusion, immune privilege, and the tumor microenvironment’, Science, Volume 348, Issue 6230, pp. 74–80, Copyright © 20152015 American Association for the Advancement of Science.

It is possible that, in the not-distant future, tumour biopsies can be described by a multidisciplinary read-out of cell composition, interactions, and molecules present, that can be interpreted for prognostic outcome and selection of optimal treatment protocols to promote immunological control (Bindea et al., 2013). At this stage, however, the goal should be to identify the more critical parameters lending themselves to therapeutic decisions.

To this end, the identification of cell expressing co-inhibitory molecules and their ligands in tumour and infiltrating cells (e.g. T cells expressing PD1 and Treg expressing CTLA4) and diverse cells expressing PDL1, offers some indication as to which tumours might be responsive to checkpoint inhibition with anti-CTLA4 and anti-PD1 antibodies (Mahoney et al., 2015; Smyth et al., 2016). In this arena, there is much need for improved reagents and standardized protocols.

In the same vein, there may be other less well-documented interactions that may provide biomarker clues for intervention. These could include indicators for costimulation with agonist antibodies such as TNFR family members, antibodies to Wnt ligand members (Sato et al., 2010), drugs influencing TGFβ‎ signalling, and those targeting (immunosuppressive) adenosine-generating ectonucleotidases CD39, CD73 in diverse immune and epithelial cells (Regateiro et al., 2011; Young et al., 2016), and the upregulation of these by TGFβ‎, offer further opportunities for another type of checkpoint blockade.

The overall principles underlying tumour immunity and its natural resistance are summarized in Figure 23.4 (Chen and Mellman, 2013).

Fig. 23.4 Stimulatory and inhibitory factors in the cancer-immunity cycle. The pathway used by the immune system to generate antitumour immunity (in green). Released tumour antigens are processed by dendritic cells that are licensed by ‘danger’ signals to leave the tissue and make their way to secondary lymphoid tissues. There they filter out antigen reactive lymphocytes from the circulation and immunize them to the tumour antigens. These ‘selected’ T cells migrate out of the lymphoid tissue into the circulation, some providing memory cells for future encounters, while others make their way into the tumour after traversing the local vascular endothelium. If the microenvironment does not impede their activity, they will then go on to damage tumour cells, release more tumour antigens, and enable the cycle to be perpetuated. For all stages in the cycle there are a myriad of potentially inhibitory mechanisms (examples in red) that, in the modern era, constitute potential targets for checkpoint blockade.

Fig. 23.4 Stimulatory and inhibitory factors in the cancer-immunity cycle. The pathway used by the immune system to generate antitumour immunity (in green). Released tumour antigens are processed by dendritic cells that are licensed by ‘danger’ signals to leave the tissue and make their way to secondary lymphoid tissues. There they filter out antigen reactive lymphocytes from the circulation and immunize them to the tumour antigens. These ‘selected’ T cells migrate out of the lymphoid tissue into the circulation, some providing memory cells for future encounters, while others make their way into the tumour after traversing the local vascular endothelium. If the microenvironment does not impede their activity, they will then go on to damage tumour cells, release more tumour antigens, and enable the cycle to be perpetuated. For all stages in the cycle there are a myriad of potentially inhibitory mechanisms (examples in red) that, in the modern era, constitute potential targets for checkpoint blockade.

Reproduced with permission from Chen DS and Mellman I, ‘Oncology meets immunology: the cancer-immunity cycle’, Immunity, Volume 39, pp. 1–10, Copyright © 2013 Elsevier Inc. https://www.sciencedirect.com/science/article/pii/S1074761313002963

Overriding tumour privilege: Immune intervention in cancer

Introductory remarks

It is becoming increasingly clear that many of the currently adopted chemotherapeutic drugs for cancer affect the immune system. It is not straightforward to determine how much of the therapeutic benefit derives from direct drug kill, rather than from effects on immunoregulatory mechanisms (Zitvogel et al., 2013; Belvin and Mellman, 2015), but it may be appropriate to consider both effects when optimizing drug dosing. Clearly, the pleiotropy of chemotherapeutics offers potentials for combinatorial synergies with emerging immunotherapeutic modalities.

In considering strategies involving vaccination or adoptive cell therapy the question arises as to which antigens could be targeted? In principle these could be self-antigens shared by normal tissues with attendant risks of debilitating collateral damage; self-antigens shared by normal cells whose loss would not be a health risk (e.g. melanocytes); self-antigens expressed in ontogeny but not in the mature individual (so-called oncofoetal antigens); and, of course, tumour neoantigens. Exploitation of the latter may require personalized therapies with their attendant commercial limitations. Of the various self-antigens, one category currently attracting attention for this reason are the numerous cancer testis antigens (Djureinovic et al., 2016; Li et al., 2016; Park et al., 2016), normally expressed on male germ cells, but aberrantly expressed on some tumours. Given the potentially tolerable collateral damage some may prove useful targets, especially for evolving generic strategies to enhance tumour kill.

Vaccination against non-viral cancers

The notion that one might have vaccines against defined self-antigens is predicated on the fact that central tolerance is not perfect, and that the autoantigens concerned may not be expressed on vital cells, or cells that cannot easily be renewed. Any therapeutic vaccine strategy needs to incorporate components that enable the licensing of dendritic cells, and numerous strategies for doing this have emerged (Palucka and Banchereau, 2013; Romagnoli et al., 2014; Shimizu et al., 2016). Although there are examples of some therapeutic benefit from vaccination to ‘self’ tumour antigens, no really effective therapeutic vaccine strategy has yet emerged (Melief et al., 2015; van der Burg et al., 2016).

By contrast, the more likely targets for vaccine development should be tumour neoantigens. However, these would, more often, need to be patient-specific and hence potentially expensive to develop. In the context of adoptive cell therapy, there may be a route to identify vaccine-peptides to select and expand TILs from individual patients for adoptive T-cell therapy.

Vaccination against viral cancers

Some 20% of human cancers are known to have an infectious origin. Prophylactic vaccines have emerged for only two of these (hepatitis B and HPV). Persistent infections with HPV offer a challenge for the development of a therapeutic vaccine, as the persistent virus needs to be eradicated before cancer develops. EBV vaccines would be highly desirable given the number of malignancies arising, but no useful vaccine has yet been demonstrated. Even in the case of HPV best efficacy data are provided in patients with premalignant lesions.

In short, our marked inability to overcome many infectious diseases through vaccination, highlights the huge challenges in developing efficacious prophylactic vaccines. Those challenges are likely to be greater for therapeutic vaccines, even where viral non-self-antigens are concerned. On this basis, it will be no easy matter to immunize cancer patients to eradicate their tumours. The presumption is that even if antigen-specific responses are induced, the ability of the host to mount efficacious effector responses will be limited by peripheral tolerance mechanisms of the kind discussed earlier. From our knowledge of the adaptive immune response, the best therapeutic immunization outcomes will most likely come where the vaccines promote the responses of both antigen-specific CD4+ T cells as well as CD8+ cells, given in such a way as to provide strong and sustained licensing of dendritic cells. Moreover, to ensure access of effector cells to tumour cells we will need to break open the strangleholds of the tumour microenvironment. This will undoubtedly require combinatorial therapy based on the emerging knowledge of its organization.

Therapeutic antibodies

The application of preformed antibodies as magic bullets became a reality when Kohler and Milstein (Kohler and Milstein, 1975) described the monoclonal antibody technology (Fig. 23.5). The initial optimism was not immediately realized for many reasons. Many of the initial rodent antibodies (Mabs) were poorly lytic even for circulating blood cells. Very few were able to utilize the human complement system for killing, while others were directed to antigens which were simply not disposed to effective lysis by cell mediated Fc dependent killing mechanisms. It also became clear that rodent IgG isotypes vary enormously in recruiting natural lytic mechanisms, with mouse IgG2a and rat IgG2b subclasses being the most effective. Some antigens were easily modulated off the cell surface, and other were simply too sparse. Rodent antibodies proved immunogenic for man, meaning that they could only be used for a short time. This early pessimism led to early studies with antibody-toxin conjugates, aiming to achieve adequate kill from a short pulse of therapy, but these generated their own problems of unwanted toxicities and additional immunogenicity.

Fig. 23.5 Genetic engineering has been successful in transforming rodent antibodies into human forms that are far less immunogenic. The technologies that evolved to achieve this have provided a platform which has catalysed the therapeutic exploitation of many facets of immune function as novel cell-bound, and also, cell-free fusion proteins.

Fig. 23.5 Genetic engineering has been successful in transforming rodent antibodies into human forms that are far less immunogenic. The technologies that evolved to achieve this have provided a platform which has catalysed the therapeutic exploitation of many facets of immune function as novel cell-bound, and also, cell-free fusion proteins.

Extensive knowledge of the structure of immunoglobulin genes, and ideas on how antibody diversity was generated, rapidly enabled the application of genetic engineering technologies to create improved immunoglobulins that nature had itself never made.

Initially, chimeric antibodies having rodent variable regions and human constant regions allowed for selection of appropriate IgG isotypes (e.g. IgG1 and IgG3 for most efficient lysis; see Bruggemann et al., 1987), and diminished their degree of foreignness to man. Shortly after, Winter and his colleagues (Jones et al., 1986) demonstrated an even greater degree of humanization by creating antibodies where only the complementary determining regions remained rodent, so reducing the risk of immunogenicity even further.

These initial observations spawned a revolution in the biotechnology sector, resulting in methods to generate antigen-specific human antibodies wholly in silico, or even replacing the rodent immunoglobulin genes with human (Bruggemann et al., 1989), so that mice could be immunized to generate human antibodies from the outset.

Even with this technology, the attempts to directly kill cancer cells have been no simple matter, except perhaps for more easily accessible cancers of the blood system. More recently toxins for conjugation have been mutated to eliminate potential T and B-cell epitopes, so enabling further efforts to exploit antibody-toxin conjugates (Mazor et al., 2016). Antibodies have also been engineered to be able to deliver chemotherapeutic drugs more selectively to the tumour site.

Overall though, failure of naked antibodies to achieve an adequate kill of solid tissue malignancies comes from limitations in tumour penetration, and failure of the Mabs to recruit a sufficient innate effector mechanism (for Fc-mediated lysis). Off-target toxicities are also real issues sometimes related to the expression of the target antigen on other body cells.

Where human monoclonal antibodies have fared better has been where they have been used to block inhibitory immune function as checkpoint inhibitors (Ribas and Wolchok, 2018) or alter cell signalling (Herceptin and Erbitux; see Katsumata et al., 2006), in addition to any immune-mediated mechanisms.

Although combination therapy with other Mabs or therapeutic modalities can give improved outcomes, there remains a commercial incentive to engineer antibody forms that may be more potent by focussing multiple effector moieties, within the one antibody construct to the tumour target. This can be achieved by incorporating multiple non-immunogenic ‘self’ effector modules in fusion proteins expressing the appropriate antibody variable regions to target the tumour. Perhaps the best early examples of such engineering are bispecific antibodies having one arm directed to the cancer antigen and the other to the CD3 epsilon chain of the TCR (Staerz and Bevan, 1989). This approach theoretically allows any T cell, regardless of its own antigen specificity, to be recruited and activated to the site of the tumour with demonstrable efficacy (Staerz and Bevan, 1989; Riethmuller, 2012). Generation of intracellular antibody fragments to critical functional sites on oncogenes is proving an exciting route to the generation of novel chemical drugs blocking interactions of oncogenes with other protein partners in malignant cells (Quevedo, 2018, Cruz-Migoni.2019).

Adoptive lymphocyte therapy

The previously mentioned work of Rosenberg and colleagues using in vitro antigen expanded TILs, has demonstrated the potential of adoptive immunotherapy (Rosenberg and Restifo, 2015). Host lymphocyte depletion was important in allowing the expanded cells to mediate tumour kill.

The unpredictability of tumour antigens coupled with human MHC polymorphisms has rendered such T-cell therapy very personalized, and not ideal for commercial development, even allowing for advances that might reduce the resistance of the tumour microenvironment.

The ideal adoptive immune cell therapy would be one where the effector cells recognized and could be activated and expanded by some endogenous ligand that is reproducibly expressed on tumour cells, and few normal cells. It is this notion that is driving interest in the possible use of γδ‎ T cells and NKT cells for cancer immunotherapy (Silva-Santos et al., 2015). The natural propensity for γδ‎ T cells to be resident in tissues may allow them to access tumours and to perform more effectively than αβ‎ T cells. Consistent with this view, is the claim that a γδ‎ T-cell signature local to the tumour is the strongest correlate of overall survival (Gentles et al., 2015).

Engineering of T cells

As is discussed in other chapters of this book, there has been extensive efforts to enhance the therapeutic power of adoptively transferred T cells. There are two general categories. The first involves use of cloned antigen-specific TCR genes introduced into human T cells (Johnson et al., 2009; Perro et al., 2010). Preclinical mouse models may allow selection of the optimal TCRs for therapeutic use (Obenaus et al., 2015). By their nature these will be MHC restricted and limited to targeting ‘self’ antigens whose off-target expression runs little risk, or to neoantigens that will likely limit utility to the individual patient. The best use of such cells will likely involve sets of gene manipulations that increase their sensitivity to antigen stimulation, limit their ability to be tolerated by their cognate antigen, endow them with adhesion and chemotactic receptors that offer optimal migration to tumour tissue, enhanced lytic machinery to kill their target, and provide them with the where-with-all to be ‘suicided’, if need be. It remains unclear, at this stage then, of what sort of ‘off-the-shelf therapies’ will offer the necessary commercial incentive to developers.

The introduction of antibody variable regions into chimeric TCR-like signalling molecules in T cells (CAR-T cells; Kalos and June, 2013; Fesnak et al., 2016; Gross and Eshhar, 2016) offers a more general approach to target cell surface antigens shared by many individuals. Once again, the most useful target antigens for these are membrane-exposed differentiation antigens on cells that are accessible and easily replenished by the host.

Perhaps not surprising therefore that malignancies of the haemopoietic system are those that have proven most targetable, thus far.

A major challenge in using third party CAR-T cells is that the host will inevitably reject them. Host preconditioning of recipients to temporarily prevent rejection might provide a sufficient window for the cells to expand, deliver, and their kill, so potentially rendering CAR-T cells as off-the-shelf-therapies. Another major challenge in CAR-T and in checkpoint blockade approaches is to avoid a cytokine storm and rampant immunopathology initiations where large numbers of de-repressed, primed T cells flood the body with an excess of immune effector molecules.

Conclusion

Cancer immunology is in a challenging phase and the major requirements to building on the therapeutic successes thus far are reported in the ‘Open Questions’. Overall, there remain many challenges. Yet, the field is in far better shape than it was 20 years ago. Then, cancer immunology was deemed to be a potential graveyard for the young immunologist. How things have changed!

Acknowledgements

I am indebted to Adrian Hayday and Elizabeth Simpson for their expert comments and many helpful suggestions on the manuscript.

Take-home message

  • In recent years, advancement in the field of tumour immunology have started to produce a range of new therapeutic strategies.

  • The notion that tumours can simply ‘escape’ the immune system has changed. It is now established that some tumours can evoke and/or provide sufficiently adjuvant-like signals which, if adequately sensed, can promote immune reactivity in the host.

  • Cross talk between tumours, stroma, and immune system however can result into an inhibition of the immune response.

  • Hence one of the aims of treatment is to override this inhibition of the immune response.

Open questions

  • Establishing safer and more effective ways to license dendritic cells to achieve high quality antigen presentation. Put another way, immunologists have still not cracked the goals of achieving super-adjuvants that are safe.

  • To break open diverse types of tumour microenvironments that limit access of T cells and antibodies that could be far more tumoricidal were more of them to get to the right place.

  • The need to identify ways of stimulating and empowering host T cells and other innate cells without depending on identifying neoantigens, as discussed in relation to γδ‎ T cells and stress surveillance (Hayday, 2009).

  • Inevitably, the application of combinatorial therapies will make in-roads (Mahoney et al., 2015; Sharma and Allison, 2015; Khalil et al., 2016), but it would be valuable to have some sort of biomarker-based algorithms that might predict the best agents to put together for individual patients.

  • For therapies directed to self-antigens, it will be imperative to establish procedures to minimize autoimmune damage to normal tissues.

Further reading

Baumeister, S. H., Freeman, G. J., Dranoff, G., & Sharpe, A. H. (2016). Coinhibitory pathways in immunotherapy for cancer. Annu Rev Immunol, 34, 539–73.Find this resource:

Chen, D. S. & Mellman, I. (2013). Oncology meets immunology: the cancer-immunity cycle. Immunity, 39, 1–10.Find this resource:

Cruz-Migoni, A., Canning, P., Quevedo, C.E., et al. C. 2019. Structure-based development of new RAS-effector inhibitors from a combination of active and inactive RAS-binding compounds. Proc Natl Acad Sci, U S A.Find this resource:

Fredriksen, T., Lafontaine, L., Berger, A., et al. (2013). Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity, 39, 782–95.Find this resource:

Joyce, J. A. & Fearon, D. T. (2015). T cell exclusion, immune privilege, and the tumor microenvironment. Science, 348, 74–80.Find this resource:

Palucka, K. & Banchereau, J. (2013). Dendritic-cell-based therapeutic cancer vaccines. Immunity, 39, 38–48.Find this resource:

Prendergast, G. & Jaffee, E. (eds) (2013). Cancer Immunotherapy. Cambridge, MA: Elsevier Inc.Find this resource:

Quevedo, C.E., Cruz-Migoni, A., Bedry, N., et al. 2018. Small molecule inhibitors of RAS-effector protein interactions derived using an intracellular antibody fragment. Nat Commun, 9, 3169.Find this resource:

Rabbitts, T.H., and Stocks, M.R. 2003. Chromosomal translocation products engender new intracellular therapeutic technologies. Nat Med, 9, 383–6.Find this resource:

Rosenberg, S. A. & Restifo, N. P. (2015). Adoptive cell transfer as personalized immunotherapy for human cancer. Science, 348, 62–8.Find this resource:

Schumacher, T. N. & Schreiber, R. D. (2015). Neoantigens in cancer immunotherapy. Science, 348, 69–74.Find this resource:

Sharma, P. & Allison, J. P. (2015). Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell, 161, 205–14.Find this resource:

Silva-Santos, B., Serre, K., & Norell, H. (2015). Gammadelta T cells in cancer. Nat Rev Immunol, 15, 683–91.Find this resource:

Zitvogel, L., Galluzzi, L., Smyth, M. J., & Kroemer, G. (2013). Mechanism of action of conventional and targeted anticancer therapies: reinstating immunosurveillance. Immunity, 39, 74–88.Find this resource:

References

Anderson, M. S. & Su, M. A. (2016). AIRE expands: new roles in immune tolerance and beyond. Nat Rev Immunol, 16, 247–58.Find this resource:

Baas, M., Besancon, A., Goncalves, T., et al. (2016). TGFbeta-dependent expression of PD-1 and PD-L1 controls CD8(+) T cell anergy in transplant tolerance. Elife, 5, e08133.Find this resource:

Banchereau, J. & Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature, 392, 245–52.Find this resource:

Belvin, M. & Mellman, I. (2015). Is all cancer therapy immunotherapy? Sci Transl Med, 7, 315fs348.Find this resource:

Beutler, B. & Rehli, M. (2002). Evolution of the TIR, tolls and TLRs: functional inferences from computational biology. Curr Top Microbiol Immunol, 270, 1–21.Find this resource:

Bindea, G., Mlecnik, B., Tosolini, M., et al. (2013). Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity, 39, 782–95.Find this resource:

Bruggemann, M., Caskey, H. M., Teale, C., et al. (1989). A repertoire of monoclonal antibodies with human heavy chains from transgenic mice. Proc Natl Acad Sci U S A, 86, 6709–13.Find this resource:

Bruggemann, M., Williams, G. T., Bindon, C. I., et al. (1987). Comparison of the effector functions of human immunoglobulins using a matched set of chimeric antibodies. J Exp Med, 166, 1351–61.Find this resource:

Chambers, C. A. & Allison, J. P. (1999). Costimulatory regulation of T cell function. Curr Opin Cell Biol, 11, 203–10.Find this resource:

Chatenoud, L., Primo, J., & Bach, J. F. (1997). CD3 antibody-induced dominant self tolerance in overtly diabetic NOD mice. J Immunol, 158, 2947–54.Find this resource:

Chen, D. S. & Mellman, I. (2013). Oncology meets immunology: the cancer-immunity cycle. Immunity, 39, 1–10.Find this resource:

Chen, W. & Konkel, J. E. (2010). TGF-beta and ‘adaptive’ Foxp3(+) regulatory T cells. J Mol Cell Biol, 2, 30–6.Find this resource:

Chiller, J. M., Habicht, G. S., & Weigle, W. O. (1971). Kinetic differences in unresponsiveness of thymus and bone marrow cells. Science, 171, 813–15.Find this resource:

Djureinovic, D., Hallstrom, B. M., Horie, M., et al. (2016). Profiling cancer testis antigens in non-small-cell lung cancer. JCI Insight, 1, e86837.Find this resource:

Dunn, G. P., Old, L. J., & Schreiber, R. D. (2004). The immunobiology of cancer immunosurveillance and immunoediting. Immunity, 21, 137–48.Find this resource:

Fesnak, A. D., June, C. H., & Levine, B. L. (2016). Engineered T cells: the promise and challenges of cancer immunotherapy. Nat Rev Cancer, 16, 566–81.Find this resource:

Fridman, W. H., Pages, F., Sautes-Fridman, C., & Galon, J. (2012). The immune contexture in human tumours: impact on clinical outcome. Nat Rev Cancer, 12, 298–306.Find this resource:

Gabrilovich, D. I., Ostrand-Rosenberg, S., & Bronte, V. (2012). Coordinated regulation of myeloid cells by tumours. Nat Rev Immunol, 12, 253–68.Find this resource:

Galon, J., Angell, H. K., Bedognetti, D., & Marincola, F. M. (2013). The continuum of cancer immunosurveillance: prognostic, predictive, and mechanistic signatures. Immunity, 39, 11–26.Find this resource:

Galon, J., Fox, B. A., Bifulco, C. B., et al. (2016). Immunoscore and immunoprofiling in cancer: an update from the melanoma and immunotherapy bridge 2015. J Transl Med, 14, 273.Find this resource:

Gentles, A. J., Newman, A. M., Liu, C. L., et al. (2015). The prognostic landscape of genes and infiltrating immune cells across human cancers. Nat Med, 21, 938–45.Find this resource:

Girardi, M., Oppenheom, D. E., Steele, C. R., et al (2001). Regulation of cutaneous malignancy by gammadelta T cells. Science, 294, 605–9.Find this resource:

Goodnow, C. C. & Basten, A. (1989). Self-tolerance in B lymphocytes. Semin Immunol, 1, 125–35.Find this resource:

Gross, G. & Eshhar, Z. (2016). Therapeutic potential of T cell chimeric antigen receptors (CARs) in cancer treatment: counteracting off-tumor toxicities for safe CAR T cell therapy. Annu Rev Pharmacol Toxicol, 56, 59–83.Find this resource:

Gubin, M. M., Artyomov, M. N., Mardis, E. R., & Schreiber, R. D. (2015). Tumor neoantigens: building a framework for personalized cancer immunotherapy. J Clin Invest, 125, 3413–21.Find this resource:

Hagerling, C., Casbon, A. J., & Werb, Z. (2015). Balancing the innate immune system in tumor development. Trends Cell Biol, 25, 214–20.Find this resource:

Hammer, G. E. & Ma, A. (2013). Molecular control of steady-state dendritic cell maturation and immune homeostasis. Annu Rev Immunol, 31, 743–91.Find this resource:

Hanahan, D. & Coussens, L. M. (2012). Accessories to the crime: functions of cells recruited to the tumor microenvironment. Cancer Cell, 21, 309–22.Find this resource:

Hanahan, D. & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell, 144, 646–74.Find this resource:

Hayday, A. & Tigelaar, R. (2003). Immunoregulation by gammadelta cells. Nat Rev Immunol, 3, 233–42.Find this resource:

Hayday, A. C. (2009). Gammadelta T cells and the lymphoid stress-surveillance response. Immunity, 31, 184–96.Find this resource:

Hinrichs, C. S., & Rosenberg, S. A. (2014). Exploiting the curative potential of adoptive T-cell therapy for cancer. Immunol Rev, 257, 56–71.Find this resource:

Hori, S., Nomura, T., & Sakaguchi, S. (2003). Control of regulatory T cell development by the transcription factor Foxp3. Science, 299, 1057–61.Find this resource:

Johnson, L. A., Morgan, R. A., Dudley, M. E., et al. (2009). Gene therapy with human and mouse T-cell receptors mediates cancer regression and targets normal tissues expressing cognate antigen. Blood, 114, 535–46.Find this resource:

Jones, P. T., Dear, P. H., Foote, J., Neuberger, M. S., & Winter, G. (1986). Replacing the complementarity-determining regions in a human antibody with those from a mouse. Nature, 321, 522–5.Find this resource:

Joyce, J. A. & Fearon, D. T. (2015). T cell exclusion, immune privilege, and the tumor microenvironment. Science, 348, 74–80.Find this resource:

Kalos, M. & June, C. H. (2013). Adoptive T cell transfer for cancer immunotherapy in the era of synthetic biology. Immunity, 39, 49–60.Find this resource:

Katsumata, M., Drebin, J. A., & Greene, M. I. (2006). Trastuzumab in breast cancer. N Engl J Med, 354, 640–4; author reply 640–4.Find this resource:

Khalil, D. N., Smith, E. L., Brentjens, R. J., & Wolchok, J. D. (2016). The future of cancer treatment: immunomodulation, CARs and combination immunotherapy. Nat Rev Clin Oncol, 13, 394.Find this resource:

Kohler, G. & Milstein, C. (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature, 256, 495–7.Find this resource:

Kurosawa, Y. & Tonegawa, S. (1982). Organization, structure, and assembly of immunoglobulin heavy chain diversity DNA segments. J Exp Med, 155, 201–18.Find this resource:

Li, B., Li, T., Pignon, J. C., et al. (2016). Landscape of tumor-infiltrating T cell repertoire of human cancers. Nat Genet, 48, 725–32.Find this resource:

Linnemann, C., Mezzadra, R., & Schumacher, T. N. (2014). TCR repertoires of intratumoral T-cell subsets. Immunol Rev, 257, 72–82.Find this resource:

Mahoney, K. M., Rennert, P. D., & Freeman, G. J. (2015). Combination cancer immunotherapy and new immunomodulatory targets. Nat Rev Drug Discov, 14, 561–84.Find this resource:

Marvel, D. & Gabrilovich, D. I. (2015). Myeloid-derived suppressor cells in the tumor microenvironment: expect the unexpected. J Clin Invest, 125, 3356–64.Find this resource:

Matzinger, P. (1994). Tolerance, danger, and the extended family. Annu Rev Immunol, 12, 991–1045.Find this resource:

Mazor, R., Onda, M., & Pastan, I. (2016). Immunogenicity of therapeutic recombinant immunotoxins. Immunol Rev, 270, 152–64.Find this resource:

McGranahan, N., Furness, A. J., Rosenthal, R., et al. (2016). Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science, 351, 1463–9.Find this resource:

Medzhitov, R. & Janeway, C. A., Jr. (1997). Innate immunity: impact on the adaptive immune response. Curr Opin Immunol, 9, 4–9.Find this resource:

Melief, C. J., van Hall, T., Arens, R., Ossendorp, F., & van der Burg, S. H. (2015). Therapeutic cancer vaccines. J Clin Invest, 125, 3401–12.Find this resource:

Mellman, I. (2013). Dendritic cells: master regulators of the immune response. Cancer Immunol Res, 1, 145–9.Find this resource:

Mitchell, G. F. & Miller, J. F. (1968). Cell to cell interaction in the immune response. II. The source of hemolysin-forming cells in irradiated mice given bone marrow and thymus or thoracic duct lymphocytes. J Exp Med, 128, 821–37.Find this resource:

Mlecnik, B., Bindea, G., Kirilovsky, A., et al. (2016). The tumor microenvironment and Immunoscore are critical determinants of dissemination to distant metastasis. Sci Transl Med, 8, 327ra326.Find this resource:

Molon, B., Ugel, S., Del Pozzo, F., et al. (2011). Chemokine nitration prevents intratumoral infiltration of antigen-specific T cells. J Exp Med, 208, 1949–62.Find this resource:

Mueller, D. L. (2010). Mechanisms maintaining peripheral tolerance. Nat Immunol, 11, 21–7.Find this resource:

Munoz-Fontela, C., Mandinova, A., Aaronson, S. A., & Lee, S. W. (2016). Emerging roles of p53 and other tumour-suppressor genes in immune regulation. Nat Rev Immunol, 16, 741–50.Find this resource:

Obenaus, M., Leitao, C., Leisegang, M., et al. (2015). Identification of human T-cell receptors with optimal affinity to cancer antigens using antigen-negative humanized mice. Nat Biotechnol, 33, 402–7.Find this resource:

Palucka, K. & Banchereau, J. (2013). Dendritic-cell-based therapeutic cancer vaccines. Immunity, 39, 38–48.Find this resource:

Park, T. S., Groh, E. M., Patel, K., Kerkar, S. P., Lee, C. C., & Rosenberg, S. A. (2016). Expression of MAGE-A and NY-ESO-1 in primary and metastatic cancers. J Immunother, 39, 1–7.Find this resource:

Perro, M., Tsang, J., Xue, S. A., et al. (2010). Generation of multi-functional antigen-specific human T-cells by lentiviral TCR gene transfer. Gene Ther, 17, 721–32.Find this resource:

Quail, D. F. & Joyce, J. A. (2013). Microenvironmental regulation of tumor progression and metastasis. Nat Med, 19, 1423–37.Find this resource:

Ramsdell, F. & Ziegler, S. F. (2014). FOXP3 and scurfy: how it all began. Nat Rev Immunol, 14, 343–9.Find this resource:

Regateiro, F. S., Howie, D., Nolan, K. F., et al. (2011). Generation of anti-inflammatory adenosine by leukocytes is regulated by TGF-beta. Eur J Immunol, 41, 2955–65.Find this resource:

Ribas, A. & Wolchok, J. D (2018). Cancer immunotherapy using checkpoint blockade. Science, 359, 1350–5.Find this resource:

Richards, D. M., Kyewski, B., & Feuerer, M. (2016). Re-examining the nature and function of self-reactive T cells. Trends Immunol, 37, 114–25.Find this resource:

Ridge, J. P., Di Rosa, F., & Matzinger, P. (1998). A conditioned dendritic cell can be a temporal bridge between a CD4 + T-helper and a T-killer cell. Nature, 393, 474–8.Find this resource:

Riethmuller, G. (2012). Symmetry breaking: bispecific antibodies, the beginnings, and 50 years on. Cancer Immun, 12, 12.Find this resource:

Rizvi, N. A., Hellmann, M. D., Snyder, A., et al. (2015). Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science, 348, 124–8.Find this resource:

Romagnoli, G. G., Zelante, B. B., Toniolo, P. A., Migliori, I. K., & Barbuto, J. A. (2014). Dendritic cell-derived exosomes may be a tool for cancer immunotherapy by converting tumor cells into immunogenic targets. Front Immunol, 5, 692.Find this resource:

Rosenberg, S. A. & Restifo, N. P. (2015). Adoptive cell transfer as personalized immunotherapy for human cancer. Science, 348, 62–8.Find this resource:

Sato, N., Yamabuki, T., Takano, A., et al. (2010). Wnt inhibitor Dickkopf-1 as a target for passive cancer immunotherapy. Cancer Res, 70, 5326–36.Find this resource:

Schaefer, L. (2014). Complexity of danger: the diverse nature of damage-associated molecular patterns. J Biol Chem, 289, 35237–45.Find this resource:

Schumacher, T. N. & Schreiber, R. D. (2015). Neoantigens in cancer immunotherapy. Science, 348, 69–74.Find this resource:

Sharma, P. & Allison, J. P. (2015). Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell, 161, 205–14.Find this resource:

Shimizu, K., Yamasaki, S., Shinga, J., et al. (2016). Systemic DC activation modulates the tumor microenvironment and shapes the long-lived tumor-specific memory mediated by CD8 + T cells. Cancer Res, 76, 3756–66.Find this resource:

Silva-Santos, B., Serre, K., & Norell, H. (2015). Gammadelta T cells in cancer. Nat Rev Immunol, 15, 683–91.Find this resource:

Simpson, E. (2006). A historical perspective on immunological privilege. Immunol Rev, 213, 12–22.Find this resource:

Simpson, E. & Nehlsen, S. L. (1971). Prolonged administration of antithymocyte serum in mice. II. Histopathological investigation. Clin Exp Immunol, 9, 79–98.Find this resource:

Smyth, M. J., Ngiow, S. F., Ribas, A., & Teng, M. W. (2016). Combination cancer immunotherapies tailored to the tumour microenvironment. Nat Rev Clin Oncol, 13, 143–58.Find this resource:

Staerz, U. D. & Bevan, M. J. (1989). Redirecting the cellular immune response. Int Rev Immunol, 4, 159–73.Find this resource:

Stein-Streilein, J. & Streilein, J. W. (2002). Anterior chamber associated immune deviation (ACAID): regulation, biological relevance, and implications for therapy. Int Rev Immunol, 21, 123–52.Find this resource:

Steinman, R. M., Hawiger, D., & Nussenzweig, M. C. (2003). Tolerogenic dendritic cells. Annu Rev Immunol, 21, 685–711.Find this resource:

Topalian, S. L., Wolchok, J. D., Chan, T. A., et al. (2015). Immunotherapy: The path to win the war on cancer? Cell, 161, 185–6.Find this resource:

van der Burg, S. H., Arens, R., Ossendorp, F., van Hall, T., & Melief, C. J. (2016). Vaccines for established cancer: overcoming the challenges posed by immune evasion. Nat Rev Cancer, 16, 219–33.Find this resource:

Vantourout, P. & Hayday, A. (2013). Six-of-the-best: unique contributions of gammadelta T cells to immunology. Nat Rev Immunol, 13, 88–100.Find this resource:

Waldmann, H., Howie, D., & Cobbold, S. (2016). Induction of immunological tolerance as a therapeutic procedure. Microbiol Spectr, 4(4). doi: 10.1128/microbiolspec.MCHD-0019-2015.Find this resource:

Wang, L., Yi, T., Zhang, W., Pardoll, D. M., & Yu, H. (2010). IL-17 enhances tumor development in carcinogen-induced skin cancer. Cancer Res, 70, 10112–20.Find this resource:

Ward, J. P., Gubin, M. M., & Schreiber, R. D. (2016). The role of neoantigens in naturally occurring and therapeutically induced immune responses to cancer. Adv Immunol, 130, 25–74.Find this resource:

Willcox, C. R., Pitard, V., Netzer, S., et al. (2012). Cytomegalovirus and tumor stress surveillance by binding of a human gammadelta T cell antigen receptor to endothelial protein C receptor. Nat Immunol, 13, 872–9.Find this resource:

Woo, S. R., Corrales, L., & Gajewski, T. F. (2015). Innate immune recognition of cancer. Annu Rev Immunol, 33, 445–74.Find this resource:

Yamabuki, T., Takano, A., Hayama, S., et al. (2007). Dikkopf-1 as a novel serologic and prognostic biomarker for lung and esophageal carcinomas. Cancer Res, 67, 2517–25.Find this resource:

Young, A., Ngiow, S. F., Barkauskas, D. S., et al. (2016). Co-inhibition of CD73 and A2AR adenosine signaling improves anti-tumor immune responses. Cancer Cell, 30, 391–403.Find this resource:

Zitvogel, L., Galluzzi, L., Smyth, M. J., & Kroemer, G. (2013). Mechanism of action of conventional and targeted anticancer therapies: reinstating immunosurveillance. Immunity, 39, 74–88.Find this resource: