Adaptive immunity programmes in breast cancer

Summary

The role of the immune system in shaping cancer development and patient prognosis has recently become an area of intense focus in industry and academia. Harnessing the adaptive arm of the immune system for tumour eradication has shown great promise in a variety of tumour types. Differences between tissues, however, necessitate a greater understanding of the adaptive immunity programmes that are active within each tumour type. In breast cancer, adaptive immune programmes play diverse roles depending on the cellular infiltration found in each tumour. Cytotoxic T lymphocytes and T helper type 1 cells can induce tumour eradication, whereas regulatory T cells and T helper type 2 cells are known to be involved in tumour‐promoting immunosuppressive responses. Complicating these matters, heterogeneous expression of hormone receptors and growth factors in different tumours leads to disparate, patient‐specific adaptive immune responses. Despite this non‐conformity in adaptive immune behaviours, encouraging basic and clinical results have been observed that suggest a role for immunotherapeutic approaches in breast cancer. Here, we review the literature pertaining to the adaptive immune response in breast cancer, summarize the primary findings relating to the breast tumour's biology, and discuss potential clinical immunotherapies.

Introduction

Breast cancer is the most common tumour type observed in American women with 246 600 new cases of breast cancer predicted for 2016.1 This high‐prevalence disease is also extremely heterogeneous, necessitating diverse treatment options that often target specific growth factors or receptors on the patient's tumour. Many of these typical treatment regimens target misregulated proliferative pathways that historically have been associated with poor patient prognoses.2 Yet, despite the treatment options currently available, breast cancer remains the second‐leading cause of cancer death of women in the USA,3 highlighting a need to find alternative therapeutic approaches. The immune system has emerged as one promising option, with immunotherapeutic approaches that uniquely harness the body's own immune system to eradicate tumours showing increased promise in the clinic.4

This review will discuss approaches that have expressly focused on the adaptive arm of the immune system, specifically the populations of specialized T cells that are capable of targeting and eliminating tumour cells when correctly signalled and mobilized. As with the rapid adoption of any new technology, the recent escalation of immunotherapeutic approaches has created an urgent need for greater foundational knowledge regarding how the adaptive immune system functions in the context of breast cancer. For this review we will: (i) describe the mechanisms by which the adaptive immune system detects and infiltrates breast neoplasms, (ii) highlight what is known regarding the cellular immune response and how it shapes breast cancer development, (iii) characterize the differences in adaptive immune response between breast cancer subtypes, and (iv) conclude with a discussion of current therapeutic approaches that modulate the adaptive immune system.

Mechanisms by which the adaptive immune system detects and infiltrates breast neoplasms

Broadly speaking, the cells involved in the immune response are part of either the innate or the adaptive immune system. Cells of the innate immune system are involved in the early response to a pathogenic threat. These cells treat pathogens in a generic manner, and limit an infection through processes such as immune cell recruitment, complement cascade, release of toxic substances, and phagocytosis. The adaptive immune system is made up of T and B cells that each express unique antigen‐specific receptors that have been formed through somatic DNA recombination and so are specialized for individual antigens. This large number of receptors increases the probability that a small fraction of these cells will recognize and proliferate in response to a specific antigen, making the adaptive immune system highly specific.5 Understanding these cell types, and their interplay of roles in the context of the tumour, can provide deeper insights both into the mechanisms by which immune infiltration occurs and the determination of individualized immunotherapeutic approaches.

Solid tumours, including breast tumours, are highly complex mixtures of both neoplastic cells and normal cell types, including stromal fibroblasts and immune cells. This community of cells exerts distinct influences upon each other, with immune cells capable of restraining or supporting tumour growth, and tumour cells capable of co‐opting immune function to their own advantage, including to suppress the antitumour immune responses.6 The immune response to a solid tumour begins in the early stages of tumorigenesis, where rapidly proliferating oncogenic cells exhaust the nutrients at the site of origin and express chemokines, such as vascular endothelial growth factor, and proteins of the C‐X‐C motif chemokine family, to promote angiogenesis. This stress induces recruitment of generally immature myeloid cells of the innate immune system to the area, resulting in a tissue repair response similar to that which occurs in a sterile wound.7 These myeloid cells are typically immunosuppressive due to the lack of an immunogenic stimulus in the tumour microenvironment. However, when this stimulus is provided, these myeloid cells begin phagocytosing and processing tumour proteins for presentation to T cells of the adaptive immune system at the lymph node.

Tumours express two classes of antigenic proteins that can be recognized by the adaptive immune system. The first are tumour‐associated antigens (TAA), which are generally normal proteins expressed in an unusual tissue or at an unusual developmental stage. Examples of TAAs include tissue differentiation antigens, cancer/testis antigens, and normal proteins that are over‐expressed in cancer.8 Antigens from oncogenic tumour viruses are sometimes considered to be TAAs as well. In addition to TAAs, somatic mutations present in oncogenic cells can alter some proteins to become neoantigens, which are unique, antigenic peptides to which the immune system has not previously been exposed.9 Only a minority of these novel peptides are efficiently processed and presented by antigen‐presenting cells and able to stimulate a strong anti‐tumour immune response by T cells. If there is sufficient immunogenic stimulus in the tumour, the neoantigens are taken up by dendritic cells that migrate to the lymph node. At the lymph node, these cells present their tumour‐derived antigens to cytotoxic T lymphocytes (CTLs) and CD4⁺ T cells, stimulating them to expand into antigenically committed clonal populations that migrate in search of the producers of their associated antigen.10 Upon identification, these T cells perform effector functions in an attempt to eliminate the neoplasm. If successful, this results in clearance of the antigen‐positive tumour subpopulation and a reduction in tumour burden. However, antigen presentation to T cells in the lymph nodes is strongly affected by signals to dendritic cells in the tumour microenvironment, some of which interfere with this process. In response to these signals, immature dendritic cells and other myeloid cells may instead create a tolerogenic environment through cell signalling and the expression of immunosuppressive cytokines. These cytokines inhibit antigen presentation and recruit or generate immunosuppressive regulatory T (Treg) cells that respond to antigens in the tumour but suppress the cytolytic activity of other immune cells, protecting the tumour from immune attack.11, 12 As a result, the immunosuppressive tumour microenvironment suppresses the proliferation and activation of CD8⁺ CTLs and helper CD4⁺ T cells that would otherwise be involved in the elimination of the neoplasm.

In breast cancer, the interplay between the innate and adaptive immune systems has begun to be characterized; however, a void in knowledge remains. The presence of adaptive immune infiltrate in breast tumours has been described in numerous studies dating back to early immunohistologically based reports in the 1980s that described malignant tumours as having various levels of T‐cell infiltration.13, 14, 15 The presence of T cells implied that at least some breast tumours were immunogenic, and recent studies have begun to elucidate what drives this immunogenicity. A role for dendritic cells has been demonstrated through numerous studies, suggesting an antigen‐dependent mechanism.16, 17, 18, 19, 20 Breast‐cancer‐specific TAAs have been identified, including human epidermal growth factor receptor 2 (HER2), mucin 1 (MUC1), carcinoembryonic antigen (CEA), human telomerase reverse transcriptase (hTERT), and Wilms tumour gene (WT1), all of which are normal proteins that have been shown to exhibit cancer‐specific over‐expression.8 Additionally, neoantigens have been implicated in driving the adaptive immune response,21 despite the low mutation occurrence in breast cancer relative to other tumour types.22 The presence of these neoantigens has been associated with increased density of T cells in these tumours and improved patient survival, indicating a potential link between neoantigen prevalence, antitumour immune response and survival.21

The presence of adaptive immune infiltration at the breast tumour indicates an active clinical immune response, presumably as an attempt to eradicate the tumour. However, despite this infiltration, the tumour often remains, suggesting an incomplete or inefficient immune response. Furthermore, local cytolytic activity, as inferred through expression of the effector‐associated genes GZMA and PRF1, is no different in breast tumours compared with normal breast tissue, suggesting that immune cells present at the tumour site may not be adequately performing their effector functions.23 There are numerous factors that may explain why this infiltration fails to translate into tumour eradication (Fig. 1). One factor is the balance of cytotoxic and inhibitory cellular infiltration, as a large number of suppressive immune cells recruited by the tumour can overwhelm the cytotoxic immune response.24 The location of the infiltration is another factor, with several studies demonstrating that immune infiltration is primarily found at the tumour periphery or within the tumour stroma where they are less likely to be successful at eradicating the tumour.13, 25, 26 However, high levels of cytotoxic T lymphocytes have also been noted at the tumour nest.13, 27 Additionally, immune checkpoint proteins expressed on effector T cells, including programmed cell death protein 1 (PD‐1; CD279) and CTLA‐4 (CD152), have been implicated in the inhibition of the host immune response to breast cancer.26, 28 These T‐cell proteins interact with surface receptors on tumours or suppressive immune cells, preventing them from functioning. Finally, immunoediting, leading to loss of antigen expression29 or HLA expression30 has been shown to protect tumours from immune‐mediated detection and elimination. Understanding the role of each of these factors in breast cancer as a whole, as well as in a patient‐specific manner, will lead to new avenues by which a patient's adaptive immune system can be harnessed for immunotherapeutic purposes.

Figure 1.

Evasion of T‐cell immune responses in breast cancer. Upon recognition of a tumour antigen, cytotoxic T lymphocytes (CTLs) clonally expand and move through the body in search of the associated antigen‐producing tumour cells to eradicate. This process is not always successful, as the tumour may evade the T‐cell response in a variety of ways including: (i) inappropriate localization in the breast cancer microenvironment, where the CTL is not capable of reaching the antigen‐producing tumour cell due to localization in the stroma or other region; (ii) immunoediting, which leads to a loss of antigen or HLA expression preventing the CTL from detecting the tumour cell; (iii) recruitment of immunosuppressive cells that express inhibitory ligands capable of suppressing CTL function; or (iv) expression of death ligands that can deactivate or kill otherwise functional T cells.

The cellular immune response and how it shapes breast cancer development

Breast tumours are infiltrated by a variety of adaptive immune cell types that uniquely shape the development and evolution of the neoplasm. Of these cell types, T lymphocytes are the best‐characterized, with several studies describing their roles in the immune response. T lymphocytes infiltrate breast tumours at varying levels, with an early study estimating that they make up between 1% and 45% of the cellular tumour mass.31 More recently, an in silico approach estimated that on average, these cells represent 26% of tumour‐associated leucocytes found in breast tumours.32 Broadly, these cells can be split up into three types: CTLs, CD4⁺ T helper cells, and immunosuppressive CD4⁺ FOXP3⁺ CD25⁺ Treg cells, as illustrated in Fig. 2. Interestingly, there have been varying reports claiming pre‐eminence for particular T‐cell subsets, with some analyses finding that CD4⁺ T cells outnumber the CD8⁺ group,33, 34 but others finding the opposite.32 It is possible that these discrepancies are due to subtype heterogeneity between breast cancer samples or differences in stromal content between patient biopsies. Understanding how these subsets of T lymphocytes interact in the context of the breast tumour microenvironment can provide a better understanding of the adaptive immune response to breast cancer, while also providing valuable prognostic information.

Figure 2.

T‐cell subsets in breast cancer. T‐cell subsets in breast cancer can broadly be divided into three categories. CD8⁺ cytotoxic T lymphocytes are involved in the direct lysis of antigen‐positive tumour cells. CD4⁺ T helper cells exist along an axis of differentiation. T helper type 1 (Th1) ‐polarized cells can enhance CD8⁺ cytotoxic T‐cell activity and can also activate antigen‐presenting dendritic cell function leading to increased anti‐tumour immunity. Th2 cells do the opposite, suppressing dendritic cell function, and are capable of inhibiting CD8⁺ T‐cell activity. Finally CD4⁺ FOXP3⁺ regulatory T cells suppress CD8⁺ cytotoxic T‐lymphocyte function, aiding in immune evasion by the tumour. These cells are also capable of suppressing Th1 T helper cell function, further suppressing cytotoxic function.

Cytotoxic lymphocytes

The CTLs have been characterized as tumour‐killing cells in various cancer types.10 These cells express T cell receptors that are capable of recognizing TAAs and cancer‐specific neoantigens. Upon antigen‐recognition, these cells undergo clonal expansion before moving through the body searching for the antigenic source cell population. Once this population has been identified, these cells act through the interferon‐γ, perforin and granzyme pathways to kill antigen‐expressing tumour cells.35 If successful in eliminating the antigen source, a majority of the effector CTL population dies, leaving behind a small population of CD8⁺ memory T cells that are capable of rapidly inducing a cytotoxic response upon future exposure to their committed antigen.

In breast cancer, several studies have sought to characterize the role of CTLs in the cancer's development. Infiltrates of these cells are associated with higher histological grade and basal phenotype, and inversely associated with oestrogen receptor (ER) and progesterone receptor expression.27, 36 Despite these associations with poor prognostic characteristics, a high CD8⁺ lymphocytic infiltrate has been significantly correlated with good prognosis in multiple studies.26, 27, 36 Additionally, a CD8⁺ CTL infiltrate is associated with pathologically complete response (pCR), where patients with high infiltrate levels have higher response rates to neoadjuvant therapy relative to those with lower infiltrate levels.37 Recent evidence suggests that tumour‐resident CD8⁺ memory T‐cell subpopulations may be the primary driver of good patient prognosis, whereas high levels of immature effector populations may be markers of an incomplete immune response.26 These results together suggest that CD8⁺ CTLs are heavily involved in anti‐tumour immune responses, and that these anti‐tumour responses can confer prolonged patient survival. Based on these observations, it is probable that some of the more severe subtypes of breast cancer, shown to contain high CD8⁺ infiltration, are more likely to respond to immunotherapeutic approaches that depend on the re‐stimulation of tissue‐resident CD8⁺ T lymphocytes or the induction of memory formation in CD8⁺ effector T‐cell populations. Going forward, it will be important to further characterize CTL infiltration with regards to its memory potential as well as to quantify the levels of anergic and exhausted T cells present in the breast microenvironment.

T helper lymphocytes

In contrast to CD8⁺ CTLs, CD4⁺ T helper cells lack cytotoxic effector functions. The T helper cells are notable for their roles in stimulating the adaptive immune response. In the context of the tumour, these cells have been shown to greatly enhance CTL infiltration and effector function.38, 39 However, these cells exist along an axis of differentiation, and understanding the polarity of the T helper cells present in the microenvironment can provide better information regarding their functionality. T helper type 1 (Th1) ‐polarized cells are a subgroup of T helper cells that promote anti‐tumour immunity, primarily through secretion of interferon‐γ.40 Conversely, Th2‐polarized T helper cells are involved in immunosuppression and promote T‐cell anergy and exhaustion through the production of interleukin‐4 (IL‐4), IL‐5, IL‐9, IL‐10 and IL‐13.40

Few studies have thoroughly characterized T helper functionality in breast cancer. The vanguard analysis for this topic was a comprehensive molecular profiling study of the infiltrating CD4⁺ T cells in breast cancer, which identified several subpopulations of T helper cells, including Th1 and Th2 cells, as well as less characterized groups of IL‐17‐producing Th17 cells, and follicular helper T‐cell types.41 Interestingly, this study found that follicular helper T‐cell infiltration was associated with extensive immune infiltration and increased pCR and overall survival.41 This finding is intriguing, as follicular helper T cells are usually found in lymph nodes where they are involved in stimulating B‐cell immunity and thus may be involved in activating protective B‐cell responses that have been described in previous studies.26, 42, 43, 44 Two additional studies found that CD4⁺ T helper cells cluster with mature dendritic cells in the tumour microenvironment. Interestingly, these studies found that the Th2 subpopulation is capable of acting through dendritic cells to promote early tumour development, and that this can be prevented through inhibition of IL‐13, further supporting the idea of Th2 cells as immunosuppressive in breast cancer.45, 46 In future studies, it will be important to further characterize T helper activity in the breast tumour microenvironment, as the levels and activity of different subpopulations of these cells probably play a large role in the efficacy of the adaptive immune response.

Regulatory T lymphocytes

CD4⁺ FOXP3⁺ CD25⁺ Treg cells are unique among T‐cell subtypes in that they are immunosuppressive in every context and have been shown to contribute to the immune escape of cancer.47, 48 In a normal setting, Treg cells act through immunosuppressive mechanisms to maintain immune tolerance to self‐antigens and prevent excess effector T‐cell induction and proliferation. In breast cancer, these cells are elevated in the peripheral blood relative to normal samples.49 Similarly, Treg levels are higher in both ductal carcinoma in situ and invasive ductal carcinoma tissue than in normal breast tissue, with invasive ductal carcinoma samples having the highest number of Treg cells.50 Increased Treg cell levels are associated with decreased relapse‐free survival and overall survival in invasive ductal carcinoma, and are also correlated with negative prognostic factors including high tumour grade, lymph node involvement, and ER negativity.50, 51 Interestingly, Treg cell activity is distinct from other prognostic factors in that it is able to identify patients who are at risk for relapse after 5 years.50

These associations agree with the immunosuppressive characteristics of Treg cells, and mechanistic studies have furthered this notion. One study found that Treg cells were recruited to the breast tumour by CCR4 secretion and activated by tumour‐associated antigen‐presenting dendritic cells.52 The Treg cells were found to be within close proximity to CTLs in the tumour infiltrate, suggesting immunosuppression.52 An additional study examined the association between Treg infiltration and B7‐H1 [programmed death‐ligand 1 (PD‐L1); CD274] expression.51 Treg cell development and function is dependent on B7‐H1, with Treg cells in low B7‐H1 environments or where B7‐H1 has been blocked, having decreased immunosuppressive capability.53, 54 This study found that Treg cell infiltrate was strongly correlated with B7‐H1‐expressing tumour‐infiltrating lymphocytes, and both cellular types were synergistically associated with ER negativity, high tumour grade, and extensive lymphocytic infiltration.51 Despite the immunosuppressive activity of Treg cells, a high ratio of CTLs to Treg cells in the surrounding tumour tissue is associated with improved overall survival, though this relationship is not valid at the tumour bed.24 Similarly, high levels of CTLs coupled with low levels of Treg infiltration were independently predictive of pCR in patients with breast cancer.55 Together, these studies support an immunosuppressive role of Treg cells in the breast tumour microenvironment. Elimination or inhibition of these cells may have potential therapeutic implications for patients with breast cancer.

Differences in adaptive immune response between breast cancer subtypes

Breast tumour expression of hormone receptors determines which growth factors the breast tumour will respond to, and are indicative of patient prognosis and response to targeted therapy. The most important of these receptors are the oestrogen and progesterone receptors, which bind oestrogen and progesterone, respectively. Patients positive for these receptors tend to have improved prognosis and respond well to hormonal therapy. Another important factor in breast cancer subtyping is HER2 expression. Unlike the hormonal receptors, patients with high HER2 expression (HER2⁺) have increased disease recurrence and poor prognosis.56 Hormone‐receptor‐negative and HER2‐negative breast cancers are relatively rare and associated with high tumour grade and high recurrence rates.56 There are currently few targeted therapies available for these types of breast cancer. The adaptive immune system tends to have different characteristics in each of these subtypes.

As noted above, infiltration of CTLs and Treg cells is higher in hormone‐receptor‐negative breast cancers relative to hormone‐receptor‐positive breast cancers.27, 36, 50, 51 Additionally, high HER2 expression is associated with increased CTL infiltration.36 Immune infiltrate status is associated with response to some chemotherapeutics, and a genomic analysis found that high scores in immune metagenes were predictive of pCR to anthracycline‐based neoadjuvant chemotherapy in the ER⁻/HER2⁻, HER2⁺ and ER⁺/HER⁻ subtypes.57 However, when adjusting for study, treatment and clinicopathological variables, this association only remained significant in the ER⁻/HER2⁻ and HER2⁺ subtypes. This finding was replicated in an independent study that focused on the association between gene expression signatures and patient prognosis.58 It is important to note that while these studies demonstrated the importance of immune infiltrates in different subtypes of breast cancer, they did not investigate the contributions by different immune cell subtypes. One effort to do this found that the expression of a T‐cell metagene was associated with prolonged survival time in ER⁻ and ER⁺/HER2⁺ breast cancers.59 Interestingly, FOXP3 expression, indicative of Treg infiltration, was associated with improved survival in HER2⁺ breast cancers, which is the opposite of what was expected given the immunosuppressive nature of Treg cells.60 Similarly, another cell‐focused analysis found that Treg and CD8⁺ memory T‐cell signatures were weakly associated with improved prognosis, even when adjusting for hormone receptor expression, HER2 expression, and other clinicopathological variables.26 Together, these results suggest differences in how the adaptive immune system responds to these cancer subtypes and illustrates how these different subtypes vary. Specifically, immune infiltration seems to play the largest role in survival in the ER⁻/HER2⁻ and HER2⁺ breast cancer subtypes.

One subtype of breast cancer that is of special interest in breast cancer immunology is triple‐negative breast cancer (TNBC). Triple negative breast cancer is a unique subtype of breast cancer characterized by negative estrogen and progesterone receptor expression, as well as a lack of HER2 over‐expression or amplification. It constitutes 10–20% of all breast cancers and more frequently affects younger patients and African‐American women.61 This subtype is notable for its poor prognostic phenotype and lack of targeted treatment options.62, 63 Several studies have reported high levels of T‐cell‐related gene expression in TNBC.57, 64 Many of these genes were found to be predictive of pCR and patient survival outcome in TNBC, including those related to T‐cell cytotoxicity, T‐cell receptor signalling, T helper cell signalling, and B‐cell activity.64 Another study found that high tumour infiltrating lymphocyte levels in either the tumour or stroma were associated with good prognosis in TNBC samples, but not in any other subtypes.65 These early findings, combined with the dearth of treatment options available, make TNBC a good candidate for immunotherapeutic approaches. As a proof of concept, recent findings from the phase Ib KEYNOTE‐012 clinical trial (NCT01848834) of single‐agent Pembrolizumab given to patients with advanced PD‐L1‐positive TNBC showed that among the 27 TNBC patients evaluable for antitumour activity, the overall activity was 18·5% with an acceptable safety profile.66 More research is still required, however, to elucidate the interplay between TNBC and the adaptive immune system and identify potential immunomodulatory routes to therapy. Going forward, it will be important to understand the mechanisms behind the subtype‐specific differences in immune infiltration and prognostic association.

Current therapeutic approaches that modulate the adaptive immune system

The paucity of effective generalized and targeted therapies, especially for triple‐negative breast cancer,67 has spurred the development of immune‐based approaches to treat refractory and metastatic disease. To date, the most successful immune‐based approaches rely on the infusion of antibodies that directly mediate anti‐tumour effector activity, without directly impacting or activating the patient's own immune response (i.e. passive immunotherapy). Widespread antibody therapy began in the late 1990s with US Food and Drug Administration approval of trastuzumab (Herceptin), a HER2‐blocking reagent that quickly became first‐line therapy for HER2⁺ tumours.68 Although some success has been realized with blocking antibodies, the field's attention has recently been focused on CD8⁺ T‐cell‐based therapies, including active vaccination,69 adoptive cell transfer,70 and immune modulating therapies (including checkpoint blockade inhibitors).71 An overview of these approaches is illustrated in Fig. 3.

Figure 3.

T‐cell immunotherapy in breast cancer. Induction and maintenance of breast cancer‐targeted T‐cell immunity remains at the forefront of pre‐clinical and clinical development. Major approaches to treat breast cancers include: (a) Active Vaccination, in which antigenic tumour‐associated antigens are delivered (in the context of autologous dendritic cells, as peptide in adjuvant, or via a viral vector) to the patient, stimulating the in situ expansion of tumour‐reactive CD8⁺ T cells; (b) Adoptive Cell Transfer (ACT), in which tumour antigen‐specific T cells are either (i) selected and expanded in vitro for re‐infusion or (ii) generated from bulk CD8⁺ populations by genetic modification to induce the expression of tumour‐specific T‐cell receptors or chimeric antigen receptors; (c) Bispecific T‐cell Engagers (BiTEs), which are fusions of immunoglobulin domains with specificity for tumours (via a surface‐expressed antigen) and T cells (usually via engagement of CD3), leading to physical tethering of tumours and T cells and subsequent activation of effector activity by the CD8⁺ cells, regardless of their nominal antigen specificity; and (d) Checkpoint blockade therapy, which uses humanized antibodies (such as Pembrolizumab and Nivolumab) to block the expression of death receptors (PD‐1) on lymphocytes, protecting them from the deactivation effects or death receptor ligand (PD‐L1) expression on tumour cells, thereby extending the effector activity of infiltrating T cells.

With the recognition that breast cancers can be immunogenic and express tumour‐specific and tumour‐associated antigens, active vaccination was pursued as a means to generate effective anti‐tumour CTLs, targeting a variety of antigens, including HER2, MUC1, CEA, cancer‐testes antigens and p53.72 Vaccines have been developed using peptide in adjuvant, viral expression vectors, killed antigen‐positive tumour cells, DNA and antigen‐pulsed dendritic cells. Each of these approaches has demonstrated safety and immunogenicity, i.e. the capacity to induce circulating antigen‐specific CD8⁺ T cells, but anti‐tumour efficacy and survival advantage have been variable and elusive.72 Paucity of response is probably a consequence of defects in T‐cell recruitment to the tumour compartment or suppression of infiltrating cells by the immunosuppressive tumour microenvironment, suggesting the need for co‐therapies to enhance vaccine efficacy.

To bypass the need for in situ induction of effector T cells, which can be impeded by immune tolerance and tumour‐induced immune suppression, adoptive cell transfer (ACT) therapy was developed. Initial studies of ACT using non‐specific cells, including allogeneic stem cell transplants, induced remission and enhanced survival, albeit with significant co‐morbidities – including profound graft‐versus‐host disease.73 Over the past decade, clinical efforts in ACT have focused on the use of autologous (patient‐derived) lymphocytes, in which the patient's own CD8⁺ T cells are selected for specificity to breast cancer‐specific antigens (including HER2) or associated antigens (such as MUC1, NY‐ESO and others),8 expanded in vitro, and re‐infused in anticipation of tumour infiltration and effector‐mediated cytolysis of cancer cells. In this setting, efficacy and enhanced survival have been observed in patients with disseminated metastases of the marrow,74 and ongoing efforts to target and eradicate primary lesions are underway with the development of new activation and expansion protocols and co‐therapies to enhance tumour access and immune activity.75

More recently, refined approaches to ACT have used genetic modification to generate tumour‐reactive cells from bulk cell populations by inducing expression of T‐cell receptors that recognize tumour antigens in the context of HLA, or chimeric antigen receptors that recognize tumour‐specific or associated surface proteins independent of HLA.76 These genetic redirection approaches potentially enable the use of autologous cells, without the difficulty and expense of in vitro selection and expansion of rare endogenous tumour antigen‐specific T cells. Ongoing clinical experience suggests that these approaches may be safe, although efficacy remains to be assessed. Genetic modification also offers the potential to impact T‐cell persistence in situ, enhance T‐cell trafficking to tumours (through the induction of tumour tissue‐matched integrins or chemokine receptors), and equip the T cells to resist immunosuppressive mechanisms in the tumour microenvironment.

A novel approach to the activation of anti‐tumour T cells is the recent development of bispecific T‐cell engagers,77 consisting of tumour‐cell‐specific and T‐cell (CD3) ‐specific immunoglobulin fragments joined by a linker; this construct physically links tumour and T cells, regardless of T‐cell antigen specificity, and this physical linkage ultimately activates T‐cell effector activity. In early trials, a HER2/CD3 bispecific construct demonstrated safety and tumour targeting,78 providing the rationale for clinical efficacy trials.

In spite of the immunogenicity and safety of vaccine and ACT approaches, clinical success has been limited, probably because of the multiple tumour‐induced T‐cell suppressive mechanisms that operate in the breast cancer microenvironment. Multiple mechanisms prevent CTL activity and promote cancer survival and spread, including catabolism of extracellular ATP or essential amino acids, infiltration of immunosuppressive cells, and tumour expression of T‐cell inhibitory ligands (such as PD‐L1). Recent studies have demonstrated that inhibition of these suppressive mechanisms can enhance the persistence, activity and efficacy of T‐cell‐based immunotherapies,79 suggesting the potential of multimodal immunotherapy approaches for treatment of breast cancers. For example, inhibition of tryptophan catabolism by indoleamine 2,3 dioxygenase80 and antibody‐mediated blockade of PD‐181 are being evaluated in conjunction with vaccines and standard of care therapy, among a variety of other approaches to suppress local tumour immunosuppression. These combinatorial approaches offer the potential to enhance T‐cell‐mediated targeting and eradicate breast cancers, including disseminated metastases.

Conclusions

The adaptive immune system has been a fertile ground for recent therapeutic advances in the treatment of some types of cancer. This part of the two‐arm system can unleash a bevy of specialized cells that destroy tumour cells when effectively stimulated. The clinical challenge has been to identify the unique immune deficits that patients with poor prognoses have and rectify that therapeutically through monotherapies or combinatorial therapies. The lack of hitherto clinical success in treating breast cancer with immunotherapy underscores the immense opportunity for discovery and development of actionable adaptive immune‐modulating targets. By better understanding how the adaptive immune system functions in the context of a patient's disease, we can develop new insights into personalized immunotherapeutic approaches that uniquely harness the body's defence mechanisms for tumour eradication.

Disclosures

The authors declare no competing financial interests.