The tumor microenvironment consists of malignant and nontransformed cells and the associations between them. Nonmalignant cells exhibit tumor-promoting functions during all stages of carcinogenesis. Targeting nonmalignant cells and/or communication mediators may have immunotherapeutic application across various tumor types and could complement other therapies as well.
1Targeting nonmalignant cells and/or communication mediators may have immunotherapeutic application across various tumor types and could complement other therapies as well.2-4
The TME is composed of a heterogeneous cell population that includes malignant cells and cells that support tumor proliferation, invasion, and metastatic potential though extensive crosstalk. Tumor-associated stromal cell populations, such as vascular endothelial cells, mesenchymal stromal cells, pericytes, fibroblasts, and adipocytes, are components of the TME. Resident cells of the TME express proangiogenic growth factors, proinflammatory cytokines, and matrix molecules, facilitating a dynamic tumor-promoting signaling network.5Immune-infiltrating cells of innate and adaptive immunity generate an immunosuppressive microenvironment to facilitate cancer progression.
“The nonmalignant cells of the TME can comprise >50% of the mass of primary tumors and their metastases, but there are still many unanswered questions regarding their biology and function,” noted Frances R. Balkwill, OBE, of the Barts Cancer Institute, and colleagues in the Journal of Cell Science.6
The innate immune system rapidly responds to stimuli to spur long-lasting antigen-specific immunity, whereas the adaptive immune system is characterized by diversity, specificity, and memory5 (the adaptive immune system will be explained in further detail during a presentation by Jonathan Powell, MD, PhD, during the Primer). Innate immunity components include skin and mucosal membranes; pattern recognition receptors (PRRs), and humoral components including complement proteins and collectins. Innate effector cell types include monocytes, macrophages, dendritic cells (DCs), natural killer (NK) cells, and neutrophils. The activation of PRRs leads to a release of proinflammatory cytokines, contributing to dendritic cell maturation and antigen presentation to the adaptive immune system.6
Antineoplastic treatment may lead to the promotion of tumor-cell apoptosis, inflammation, and tumor-associated antigen (TAA) presentation with antitumor T-cell immunity activation.5In growing tumors, persistent hypoxia upregulates the production of high mobility group box 1 (the HMGB1 protein), promoting angiogenesis and tumor growth by recruiting endothelial precursor cells and suppressive tumor-associated macrophages (TAMs).7TAMs in the TME are characterized by the activated M2 phenotype and promote tumor cell growth and survival via immune suppression and tumor vascularization.8,9
As part of the innate immune system, the complement system is thought to have protective effects through both direct complement activation and tumor-antibody directed complementdependent cytotoxicity. In contrast, C5a generation in the TME attracts myeloid-derived suppressor cells (MDSCs) and induces reactive oxygen and nitrogen species generation, leading to impaired tumor-directed T-cell effects.10
“There is evidence that both acute and chronic inflammation may promote genetic abnormalities and cancer progression, suggesting that modulation of the innate immune system has the potential to impact our approach to cancer immunotherapy,” explained David B. Page, MD, of Memorial Sloan Kettering Cancer Center, and colleagues, in the Journal for ImmunoTherapy of Cancer.5
Many different populations of T cells are found within the TME, which is the focus of a presentation by Ana C. Anderson, PhD. Cytotoxic CD8-positive memory T cells are generally antigen-experienced, have tumor cell killing functions, and have been strongly associated with a good prognosis.11CD4-positive T helper 1 (TH1) cells support these CD8-positive T cells, and have been associated with a good prognosis when highly abundant within in the TME. CD4-positive TH2 cells and TH17 cells, which support B-cell responses and inflammation, respectively, have been associated with good outcomes in breast and esophageal cancers.12,13
Immunosuppressive T-regulatory cells (Tregs) are the CD4-positive T cells most frequently associated with tumor promotion.14 Through the production of interleukin-10, transforming growth factor beta (TGF-β), and cell-mediated contact through cytotoxic T-lymphocyte antigen-4 (CTLA-4), Tregs inhibit the immune system’s ability to recognize and clear tumor cells.14,15A worse prognosis has been associated with an increased presence of Tregs in the TME in many cancer types.16-18Their presence, however, has been associated with a good prognosis in Hodgkin lymphoma, which is attributed to direct tumor cell growth suppression.19-21
During T-cell activation, positive costimulatory receptors, including CD40, OX40, and CD137, are upregulated, allowing for differentiation into memory T cells and cytokine polarization.22Inhibitory receptors, such as programmed death-1 (PD-1) and CTLA-4, are induced following antigen encounters to limit immune system overstimulation. CTLA-4 is induced on CD4-positive T cells and CD8-positive T cells early after activation and is constitutively expressed on Tregs.23Additionally, binding of CTLA-4 inhibits T-cell proliferation. PD-1 is expressed on activated NK cells, T cells, and B cells. PD-1 binds to its ligand, PD-L1, which is expressed by tumor and immune cells. B cells are commonly found in draining lymph nodes and lymphoid structures adjacent to the TME, but can also be found at tumor invasive margins. In some breast and ovarian cancers, B-cell infiltration into the TME has been associated with a favorable prognosis.24,25
NK cells and natural killer T (NKT) cells are innate cytotoxic lymphocytes that infiltrate the tumor stroma but do not come into direct contact with tumor cells. The presence of NK and NKT cells has been indicative of a good prognosis in colorectal and many other cancers.26 NK cells may not be able to exert their tumor-killing function, however, due to the induction of an anergic phenotype by TGF-β derived from malignant cells.11
Pro-tumorigenic activities of TAMs have been reported in most cancer types.27TAMs have an important role in malignant cell migration, invasion, and metastases.28The presence of TAMs in the microenvironment and the expression of genes associated with a strong macrophage signal are both correlated with a poor prognosis.29,30The interaction between macrophages and the TME determines phenotypic response to environmental conditions. Many TAMs accumulate in hypoxic and necrotic areas of tumors as a result of hypoxia-induced chemoattractants, such as vascular endothelial growth factor.31MDSCs are defined as inhibitory immune cells present in several cancer types. Because of a high degree of phenotypic variability and differentiation potential, characterization of MDSCs is complex.32MDSCs have been shown to inhibit CD8-positive T-cell activation, induce Treg development, and modulate the polarization of macrophages to a TAM-like phenotype.33-36
DCs, which are important for antigen processing and presentation, and the topic of an early Primer presentation by A. Karolina Palucka, MD, PhD, are defective and unable to sufficiently stimulate an immune response to TAAs if they are found in the TME.36 The ability of DCs to activate immune functions is further impaired by the hypoxic and inflammatory TME. Furthermore, some DCs have been shown to suppress T-cell responses at the tumor site. Residential fibroblasts differentiate into myofibroblasts in response to tissue injury, which can cause organ fibrosis and enhance cancer development risk.37,38Many TMEs contain large numbers of myofibroblasts, also known as cancer-associated fibroblasts (CAFs), which have many precursors, including smooth muscle cells, mesenchymal stem cells, and endothelial cells.39,40CAFs secrete fibroblast growth factors, and in turn, TGF-β from fibroblasts induce epithelial-mesenchymal transition in malignant cells promoting an immune-suppressive microenvironment.41Fibroblasts also secrete extracellular matrix components and remodeling enzymes that contribute to the composition of the TME.
How Tumor Cells Evade Immunosurveillance and Detection
“Some tumors effectively exclude the immune system,” explained Thomas Gajewski, MD, PhD, of the University of Chicago, in an interview with ScienceLife.42 “They hide from it or prevent T cells from invading. Others allow some mingling, an initial dialogue with the immune system, then suppress it. These 2 categories may require distinct immunotherapeutic interventions for optimal clinical effect.” Tumors can evade immune elimination using tumor-intrinsic and tumor-extrinsic mechanisms.5Intrinsic mechanisms include major histocompatibility complex (MHC) loss, antigen loss, secretion of immunosuppressive cytokines, or expression of cell-surface markers that alter T-cell functionality. Extrinsic factors include geographic barriers, a variety of regulatory or suppressive immune cells (including Tregs), and a heterogeneous population of MDSCs and TAMs. Tregs may be targeted and depleted through the use of inhibitory antibodies, modification of trafficking, or T-cell plasticity exploitation.43 Ipilimumab (Yervoy), an anti−CTLA-4 immunotherapy, may function partially by depleting Tregs from the TME.44
By hiding tumor-specific antigens (TSAs), cancer cells can escape the host immune system. Specifically, immune cells search for cancer neoantigens, which are produced as a result of tumor-specific nonsynonymous mutations in gene-coding regions and are cross-presented by antigen-presenting cells.45Recent studies demonstrated that tumor antigens are strongly correlated with both immunogenicity and response to anti−PD-1 immunotherapy.46-47Cancer cells can evade the immune system and immunotherapies by reducing neoantigen expression.48Various molecular events can facilitate the loss of neoantigens, including the downregulation or loss of MHC class I proteins, which was confirmed in high-throughput studies.45,49,50MHC class I proteins present small peptide antigens on cell surfaces and activate CD8-positive T cells via the T-cell receptor. T-cell epitopes associated with impaired peptide processing antigens presented by residual MHC class I molecules of cancer cells that have undergone immunoediting may be exploited using vaccination approaches.51
Other important mechanisms of immune escape include defects in TAP-1 and TAP-2 (transporter associated with antigen processing proteins) or beta-2 microglobulin that alter antigen processing.50,52Antitumor responses can also be impacted by alterations in the interferon-gamma (IFN-γ) signaling pathway.53Immune attacks may be dampened if cancer cells overexpress antiapoptotic proteins, such as Bcl-xL.54
Cancer cells produce immunoregulatory enzymes, immune checkpoint ligands, or immunosuppressive cytokines to modulate the efficacy and duration of the immune response, similarly to how MDSCs or Tregs present in the TME.55TGF-β impairs the activation, maturation, or differentiation of innate and adaptive immune cells, including CD4-positive T cells, CD8-positive T cells, NK cells, dendritic cells, neutrophils, and macrophages.56 Furthermore, TGF-β inhibits cytotoxic T-cell functionality by altering expression of cytotoxic products such as IFN-γ.57,58 TGF-β has also been shown to inhibit differentiation into central memory T cells.59
The clinical success of many novel immunotherapies targeting immune checkpoints highlights the importance of associated ligands in regulating immune response.60By expressing checkpoint molecules such as PD-L1, cancer cells can dampen immune response and prevent T-cell targeted cytotoxicity. T-cell activity is reduced upon surface-expressed PD-1 association with cognate PD-L1 molecules, eventually leading to an exhaustion state.51,61,62T-cell activity can be partially restored through the introduction of inhibitory antibodies that block immune checkpoint functionality. The inflammatory cytokines IFN-γ and tumor necrosis factor-alpha can induce PD-L1 expression to limit the immune response in an inflammatory environment. Cancer cells encounter a variety of alterations and signals can modulate PD-L1 expression as an immunosuppression mechanism.63,64
In ex vivo experiments, MDSCs directly suppressed T-cell proliferation and are associated with poor survival outcomes following treatment with ipilimumab. M2-like TAMs contribute to the shaping of the TME through the secretion of cytokines related to tissue remodeling, enhanced invasion and metastasis, and increased immune suppression.65Various compounds, including colony-stimulating factor 1 receptor, are under investigation for targeting and blocking MDSCs and TAMs; however, specificity remains a challenge.
The FDA has approved immune checkpoint inhibitors as pembrolizumab (Keytruda), nivolumab (Opdivo), and ipilimumab for treating a variety of cancer types.5“Checkpoint- blockade immunotherapy has arguably been the most exciting advance made in cancer treatment in recent years,” Antoni Ribas, MD, PhD, of the University of California, Los Angeles, wrote in a recent perspective published in the New England Journal of Medicine.66These antibodies dampen or shut down T-cell responses by binding to PD-1 or CTLA-4. Improved survival has been seen in various cancer types with these inhibitors and with combinations of inhibitors blocking both pathways.67
Cancer vaccines are designed to induce an adaptive immune response to a specific administered antigen, although unlike vaccines for preventing infectious disease, cancer vaccines are typically therapeutic.5Identifying appropriate antigen targets for generating an effective antitumor response is challenging because tumor antigens may be either tumor associated or tumor specific. TAAs are present in many tumor cells, but may also be present in some normal cells. TSAs are antigens that are only found in tumors. Currently, few patients achieve tumor regression with cancer vaccines, potentially as a result of a lack of antigen-specific T-cell development, suboptimal selection of targeted antigens, suboptimal antitumor activity of vaccine-induced T cells, suppression of T cells by tumor-derived immunosuppressive signals, or prevention of activated T cells to enter the tumor.68
Adoptive cellular therapy involves manipulating autologous T cells outside of the body and reinfusing them into a patient to generate a robust antitumor response.5T cells can be manipulated by gene insertion of a chimeric antigen receptor (CAR) or engineered T-cell receptor (TCR), or through expansion of endogenous tumor-infiltrating lymphocytes. CAR-engineered T cells can associate with tumors more readily and can be activated directly without antigen binding. Second-generation CARs have been enhanced using costimulatory domains. CAR T-cell therapy has contributed to a 90% complete remission rate in a population of primarily pediatric patients with acute lymphoblastic leukemia (ALL), which led to the approval of tisagenlecleucel (Kymriah) in this indication.69This was followed by an approval for axicabtagene ciloleucel (Yescarta) for adult patients with relapsed or refractory non-Hodgkin lymphoma.
Using gene transfer technology, TCR genes can be introduced into T cells and promote an increased affinity against tumor-specific complexes. Unfortunately, this technique has not been successful in the clinic due to cross-reaction with cardiomyocyte proteins and development of severe neurologic adverse events.70-72“Future questions in the field include how to combine adoptive T-cell therapy with other types of immunotherapy, how to scale up and streamline manufacturing, and the challenge of identifying novel targets with minimal ‘on-target, off-tumor’ toxicity,” noted Page and colleagues in the Journal for Immuno-Therapy of Cancer.5Bispecific antibodies link the variable regions of 2 antibodies, leading to a construct that is specific to 2 agents. Blinatumomab (Blincyto), an antibody specific to both CD3 and CD19, has been approved by the FDA for adult and pediatric patients with relapsed or refractory B-cell ALL.73Blinatumomab engages T cells with malignant CD19-positive leukemia clones, leading to T-cell activation and cytolysis of the leukemia cell. Understanding the interactions between the TME and immunotherapy facilitates the understanding of mechanisms of action and the development of new approaches. “As our knowledge about the TME and immunotherapy expands, more approaches will undoubtedly emerge to harness the TME and improve our current immunotherapies,” explained Haidong Tang, PhD, and colleagues of the University of Chicago, in Cancer Letters.74