Introduction

• ■
  

  The origins of immunotherapy can be traced back to the early 1900s when Paul Ehrlich postulated that the host immune system plays a role in the early recognition and elimination of malignant cells. Over the next century, many other researchers built upon this notion. Theories of immunosurveillance ^, and cancer immunoediting were conceptualized, to help explain the complex interplay between the immune system and carcinogenesis. ^, The theory of immunosurveillance proposed by Macfarlane Burnet and Lewis Thomas states that cellular immunity performs constant surveillance in the human body, and is able to identify and eliminate malignant cells in the early stages of carcinogenesis. This was followed by several experiments in murine models where the incidence of malignancies in athymic mice was studied by researchers. Unfortunately, these showed discordant results and failed to definitively show a relationship between immunosuppression and carcinogenesis. Interest in the role of immunity in carcinogenesis waned until the 1990s when new discoveries revived interest in this field.

  
  
• ■
  

  It was during this time that better understanding of the complex interactions between the immune system and malignant cells led to the evolution of the theory of immunoediting. Immunoediting is best described by the three “E’s” which represent different stages in the process. The first “E” stands for “Elimination,” in which immune cells recognize and eliminate malignant cells. The second “E” represents “Equilibrium,” in which the tumor cell variants that escape elimination reach an equilibrium with the immune system. Although the immune system continues to evolve and destroy some of these variants, new variants of tumor cells are constantly being produced resulting in incomplete elimination and an equilibrium between the immune system and the tumor cells. A variety of cytokines and chemokines such as interferon γ, CXCL10, CXCL9, and CXCL11 produced by the tumor cells and the infiltrating immune cells modulate the interactions between the two in the tumor microenvironment. The third “E” stands for “Escape,” in which the tumor variants that have evolved to evade recognition by the immune system start to multiply rapidly. This escape is mediated by multiple modalities, such as loss of major histocompatibility complex (MHC) class 1 expression; antigenic mimicry; alteration of the chemokines leading to an increase in immunosuppressant cells, such as regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs); and upregulation of the immune checkpoints leading to T cell exhaustion.

  
  
• ■
  

  Early experiments that attempted to harness host immunity to combat cancer were performed as far back as 1777 when the surgeon to the Duke of Kent tried to create a cancer vaccine by injecting himself with malignant cells. These remained unsuccessful until 1891 when William Coley reported a 10% cure rate in soft-tissue sarcomas by using inactivated streptococci and Serratia marcescens . Over the next century recognition of the key role of cytokines led to the successful application of high-dose interleukin (IL)-2 in the treatment of melanoma and renal cell carcinoma. In the last decade, the field of immunotherapy has grown in leaps and bounds with the discovery of immune checkpoint inhibitors. In contrast to high-dose IL-2 which is associated with a number of serious adverse effects, these agents are much better tolerated. They have also shown remarkable efficacy in a wide range of malignancies and are currently approved for use in patients with advanced melanoma, non–small-cell lung cancer, head and neck squamous cell cancer, classic Hodgkin’s lymphoma, urothelial carcinoma, renal cell carcinoma, and advanced microsatellite instability (MSI) high/mismatch repair deficient tumors, among others. As immunotherapeutic agents interfere with the physiological pathways that regulate immune homeostasis, a spectrum of untoward side effects that resemble autoimmune diseases have been noted with their use. These adverse events are called immune-related adverse events or irAEs. In this chapter, we will explore the potential pathophysiological mechanisms leading to immune-related adverse events. We will begin by describing the physiological pathways responsible for immune tolerance and then go on to examine how alterations in these pathways can potentially lead to autoimmunity.

Physiological Pathways Involved in the Regulation of Immune Homeostasis

• ■
  

  Both the innate and adaptive immune systems play an important role in generating immune responses in a normal host. Whereas the innate immunity is nonspecific and does not require antigen presentation by the MHC, adaptive immunity is more versatile and leads to the activation and expansion of antigen specific immune cells. Under physiological conditions, adaptive immunity is tightly regulated by costimulatory and inhibitory signals. T cell activation requires two signals—the first one is mediated by engagement of the T cell receptor with an antigenic peptide presented by MHC, and the second is mediated by binding of costimulatory molecules on T cells to ligands on antigen presenting cells. In naïve T cells, the interaction of the costimulatory CD28 with B7-1 and B7-2 (also known as CD80 and CD86, respectively) plays an important role in downstream signaling, which eventually leads to the secretion of proinflammatory cytokines such as IL-12 and interferon-γ, resulting in clonal expansion. The inhibitory signals allow for contraction of the T cell clone upon resolution of the antigenic stimulus.

  
  
• ■
  

  The diversity of the T cell receptors (TCR) and B cell receptors (BCR) is a result of the recombination of three separate gene segments—the variable (V), diversity (D), and joining (J) genes—during the differentiation of B cells and T cells in the bone marrow (B cells) and thymus (T cells), respectively. Studies have estimated that between 20% and 50% of TCRs and BCRs generated by this process can have affinity to a self-antigen ; however, only a fraction of the general population develops clinical manifestations of autoimmunity. The inhibitory signals or immune checkpoints play an important role in immune tolerance of self-antigens by downregulation of the self-reactive cells. Some of the prominent immune checkpoints include cytotoxic T-lymphocyte–associated protein 4 (CTLA-4), programmed cell death-1 (PD-1), T-cell immunoglobulin and mucin-domain containing-3 (TIM3), and lymphocyte-activation gene 3 (LAG3). Activation of T cells in the lymphoid organs leads to expression of CTLA-4, which is homologous to CD28 but binds to B7-1 and B7-2 with a much higher affinity than CD28, resulting in negative regulation of the T cell response. PD-1 plays an important role in regulation of T cell responses in peripheral tissues. PD-1 binds to its ligands PD-L1 and PD-L2 and leads to apoptosis of antigen-specific T cells and a reduction in the apoptosis of Tregs, therefore dampening the immune response.

Pathophysiology of Immune-Related Adverse Events (irAEs)

• ■
  

  Tumor cells can upregulate immune checkpoint pathways to evade the host immune system. Immune checkpoint inhibitors mediate their antitumor effects by releasing this inhibition and reinvigorating the immune responses to malignant cells. Predictably, this also interferes with the process of immune homeostasis as described previously and can result in immune-mediated adverse effects. The precise pathophysiology of irAEs is unknown; however, emerging evidence suggests that many different aspects of the immune system may play a role. In the following section, we describe the proposed mechanisms for irAEs and the evidence to support each of them.

Shared Antigens, Cross Presentation of Neoantigens and T Cell Epitope Spreading

• ■
  

  Shared antigens: Shared antigens amongst cells of various organs and the regulatory proteins that monoclonal antibodies are directed against may be responsible for some of the toxicities seen with immune checkpoint inhibitors. For example, the normal pituitary gland expresses CTLA-4, which may explain the higher incidence of immune-mediated hypophysitis in patients treated with CTLA-4 monoclonal antibodies. ^,

  
  
• ■
  

  Epitope spreading: Tumor-associated antigens that are homologous to native epitopes found on normal cells likely play a role in promoting immune tolerance to malignant cells. Epitope spreading is the process by which the immune response diversifies from targeting a specific epitope to subdominant epitopes in the target protein. Heightened immune responses with immunotherapy can target these shared or homologous antigens on tumor cells and normal tissues, leading to immune-related toxicities. In a report of two cases of fulminant myocarditis following treatment with immune checkpoint blockade, the authors found high expression of muscle specific antigens (desmin and troponin) on tumor cells, further supporting this notion. Likewise, it has been postulated that shared antigens among melanoma cells and melanocytes may explain the higher incidence of vitiligo noted in patients with melanoma treated with immunotherapy.

  
  
• ■
  

  Cross presentation of neoantigens: One of the mechanisms of action of cancer immunotherapy is by antigen spreading where T cell–mediated destruction of tumor cells leads to the release of additional tumor-associated antigens which can then generate secondary immune responses. Therefore, immunotherapy remodels the immune repertoire of circulating T cells by generation and expansion of new clones. Some of these diversified T cell clonotypes may cross react with antigens on normal tissues and result in autoimmunity. In a retrospective analysis of peripheral blood mononuclear cells PBMCs from patients with metastatic prostate cancer treated with ipilimumab, diversification of the T cell receptor (TCR) repertoire with broad expansion of low frequency clonotypes was associated with the development of irAEs. Specifically, patients who developed irAEs had a significantly greater fraction of new clones that were not present at baseline in comparison with those who did not develop irAEs. In this scenario, it is plausible that the broadening of the clonotypes of T cells rather than priming of T cells with autoantigens is responsible for tissue inflammation and irAEs. Similar findings from a phase II trial of ipilimumab and androgen deprivation therapy in patients with metastatic prostate cancer revealed a correlation between clonal expansion of CD8 T cells and subsequent development of irAEs.

Dysregulation of Other Immune Cells

• ■
  

  The balance between maintaining immune homeostasis and effective immune responses is tightly regulated by various immune cells and regulatory cytokines. Disruption of this equilibrium by immune checkpoint inhibitors can result in unrestrained immune responses, leading to irAEs. Some of the key modulators of immune responses that appear to play a role in the pathogenesis of irAEs are described below.

  
  
  
  
  • 1.
    

    Regulatory T cells: FOXP3 Tregs play a vital role in suppressing autoreactive T cells as well as the regulation of adaptive T cell responses. They are also vital in maintaining mucosal tolerance to commensal bacteria in the gastrointestinal tract. Previous studies have shown that commensal bacteria can induce the formation of adaptive Tregs and IL-10 within the gut, allowing them to persist in the gut without inducing inflammation. In fact, many preclinical models of inflammatory bowel disease highlight the association between depletion of Tregs in the gut and inflammatory bowel disease. Researchers have demonstrated that CTLA-4 is constitutively expressed on the surface of Tregs and of vital importance in the suppressive function. An increase in the ratio of effector T cells to Tregs has been observed after treatment with immunotherapy. Dysregulation of this homeostasis maintained by Tregs is postulated to play a role in some irAEs, particularly colitis. In fact, studies looking at the histopathological findings in patients with inflammatory bowel disease and colitis from immune checkpoint inhibitors have shown a similar number of mucosal Tregs in both populations. However, other studies have reported no difference in the number of mucosal Tregs between patients with and without colitis after treatment with a anti-CTLA-4 antibody.

    
    
  • 2.
    

    Eosinophils: Some patients with dermatological irAEs have been noted to have eosinophilic infiltrates similar to the histopathological findings in a variety of autoimmune skin conditions. In addition, retrospective studies have also shown a relationship between elevated eosinophil counts early in the course of treatment and subsequent development of irAEs. ^,

  
  

Autoantibodies

• ■
  

  Although a majority of irAEs are a consequence of T cell–mediated toxicity, some irAEs have been associated with altered humoral responses. A challenge with proving causality of these autoantibodies is that many patients do not have pretherapy antibody titers. Therefore it is difficult to distinguish patients with preexisting antibodies from those who develop new autoantibodies after immunotherapy. Studies have reported thyroid dysfunction associated with antithyroglobulin antibodies after treatment with immune checkpoint inhibitors. Iwama et al. demonstrated that in a murine model after administration of a CTLA-4 monoclonal antibody, pituitary antibodies were detected in those with hypophysitis. These antibodies were directed against the normally expressed CTLA-4 in pituitary cells and led to excessive complement activation and infiltration of the pituitary with lymphocytes. In checkpoint blockade–associated bullous pemphigoid dermatosis, the pathognomonic antibodies to BP180 and BP230 can be detected in the serum. In some patients with diabetes mellitus induced by checkpoint inhibitors, autoantibodies such as anti-glutamic acid decarboxylase have been detected and are thought to be central to the pathogenesis. In certain rheumatological disorders, the presence of preexisting antibodies or development of new antibodies, such as anti-cyclic citrullinated peptide, have been associated with the development of rheumatoid arthritis or other autoimmune phenomenon with therapy.

Cytokines

• ■
  

  Activation of effector T cell function by immune checkpoint inhibitors leads to a surge in proinflammatory cytokines and recruitment of inflammatory cells leading to organ damage. Cytokine release syndrome (CRS) is a potentially life-threatening adverse effect that has been observed with certain antibodies used for the treatment of leukemia and chimeric antigen receptor CAR T-cell therapy. CRS is associated with high circulating levels of many cytokines including IL-6 and interferon-γ. IL-6 appears to be central to the pathogenesis of CRS; therefore blocking this cytokine with tocilizumab (a humanized IL-6R monoclonal antibody) has proven to be an effective management strategy for this syndrome.

  
  
• ■
  

  Another downstream effect of immune dysregulation by checkpoint inhibitors is an imbalance between regulatory cytokines. Higher levels of the proinflammatory cytokine IL-17 has been associated with subsequent development of colitis in patients with melanoma treated with ipilimumab. On the other hand, a decrease in the levels of antiinflammatory cytokines such as IL-10 has been observed in patients with irAEs.

Genetic Polymorphisms

• ■
  

  The association between genetic polymorphisms in the PD-1 and CTLA-4 genes and autoimmune conditions is well established. It is plausible that some patients who go on to develop irAEs with immunotherapy have germline genetic polymorphisms in key regulatory genes in immune pathways predisposing them to irAEs. Similarly, high-risk HLA genotypes have been associated with disorders such as type 1 diabetes. There are several reports in the literature of patients with these high-risk genotypes who developed type 1 diabetes with immunotherapy. One group reported that certain germline polymorphisms in the T cell receptor beta variable gene were associated with a higher risk of irAEs. This is likely due to the altered ability of these T cell receptors to interact with HLA, leading to increased autoantigen reactivity. Another group reported that certain single nucleotide polymorphisms in genes that contribute to PD-1-associated T cell responses were associated with adverse effects with nivolumab.

Environmental Modulation of Immune Responses

• ■
  

  There is mounting evidence to suggest that environmental modulation of host immunity, particularly by the gut microbiome, is associated with both the efficacy and toxicity of immune checkpoint inhibitors. Murine models highlight the pivotal role of the gut microbiome in the maturation of the host immune system and regulation of mucosal immunity. It is therefore not surprising that the composition of the host gut microbiome can predispose them to irAEs. Studies have shown that a higher number of commensal bacteria (Bacteroides) in the fecal microbiome are associated with a lower incidence of colitis. ^, On the other hand, bacteria of the phylum Firmicutes have been associated with an increased risk of colitis.

Specific Immunotherapeutic Agents and their Associated Toxicities

• ■
  

  Cytokines

  
  
  
  
  • ■
    

    Cytokines are among the first immunotherapeutic agents to be used in clinical practice. High dose IL-2 was approved for use in patients with melanoma and renal cell carcinoma due to its ability to induce durable responses and cure in a small fraction of patients. Predictably, infusion of high-dose cytokines can lead to a cytokine storm manifesting as hypotension, cardiac arrhythmias, pulmonary edema, and fever, which can be life threatening. These are direct consequences of the unbridled inflammation induced by the cytokine storm. Some of the novel cytokine superagonists are less likely to result in a cytokine storm.

  
  
  
  
• ■
  

  Immune checkpoint inhibitors

  
  
  
  
  • ■
    

    The immune checkpoint inhibitors that are currently approved for clinical use target PD-1, PD-L1, or CTLA-4. The profile and incidence of immune-related toxicities observed with PD-1/PD-L1 inhibitors is different from those seen with CTLA-4 inhibitors. Colitis and hypophysitis are reported more commonly with CTLA-4 inhibitors, whereas pneumonitis has been reported more with PD-1 and PD-L1 inhibitors. This is likely a reflection of the role of these proteins in different stages of immune regulation—CTLA-4 regulates the priming phase of T cell activation, whereas PD-1 regulates the effector phase in peripheral tissues. Overall, CTLA-4 inhibitors have been associated with higher incidences of irAEs than have PD-1/PD-L1 inhibitors. Combined PD-1/CTLA-4 inhibition is associated with a higher incidence of irAEs than is either agent alone. The site of primary disease also appears to influence the organ involvement in irAEs. For example, pneumonitis is seen more commonly in patients with non–small-cell lung cancer treated with immunotherapy. This is probably a consequence of organ specific immune response leading to a higher grade of inflammation restricted to that site.

  
  
  
  
• ■
  

  Chimeric antigen receptor T (CAR T) cells

  
  
  
  
  • ■
    

    CAR T-cell therapy has been approved for patients with refractory acute lymphoblastic leukemia (ALL) and diffuse large B cell lymphoma (DLBCL). A very high incidence of cytokine release syndrome (CRS) is seen with CAR T-cell therapy due to supranormal levels of cytokines such as IL-6. Therapy targeting IL-6 (tocilizumab) has been successful in mitigating symptoms of CRS.

  
  
  
  
• ■
  

  Bispecific T cell engagers (BiTEs)

  
  
  
  
  • ■
    

    BiTEs are antibodies which link the CD3 molecule expressed on T cells with a specific antigen. In the case of blinatumomab, the target molecule is CD19. Blinatumomab was recently approved for use in Philadelphia chromosome-negative B-ALL and has been associated with CRS and neurological toxicities secondary to immune dysregulation.

  
  
  
  
• ■
  

  Oncolytic viruses

  
  
  
  
  • ■
    

    irAEs such as glomerulonephritis, pneumonitis, colitis, and vasculitis have been noted with the use of the oncolytic virus talimogene laherparepvec or T-VEC. It is likely that the immunostimulatory properties of this virus mediate these adverse effects.

  
  

Conclusion

Immune-mediated toxicities are the expected adverse consequences of the loss of immune tolerance with immunotherapy. As we gain more experience with the clinical use of immunotherapeutic agents, we have also gained more insight into the varied mechanisms of immune-related toxicities. Due to the complex nature of immune modulation by interactions between different entities, both within and outside of the human body, a comprehensive understanding of these immune-mediated toxicities will require in-depth study of each of those entities. A thorough understanding of the mechanisms involved will aide physicians in patient selection for immunotherapy and inform the optimal therapy for immune-mediated toxicities.