Our Current Understanding of Checkpoint Inhibitor Therapy in Cancer Immunotherapy

Abstract

Objective

Treatments with FDA-approved blocking antibodies targeting inhibitory cytotoxic T lymphocyte antigen 4 (CTLA4), programmed cell death protein 1 (PD-1) receptor or the programmed cell death ligand 1 (PD-L1), collectively named checkpoint inhibitors (CPI), have been successful in producing long-lasting remissions, even in patients with advanced stage cancers. However, these treatments are often accompanied by undesirable autoimmune and inflammatory side effects, sometimes bringing severe consequences for the patient. Rapid expansion of clinical applications necessitates a more nuanced understanding of CPI function in health and disease to develop new strategies for minimizing the negative side effects, while preserving the immunotherapeutic benefit.

Data Sources

This review summarizes a new paradigm shifting approach to cancer immunotherapy with the focus on the mechanism of action of immune checkpoints (CTLA4, PD-1 and its ligands).

Study Selections

We performed a literature search and identified relevant recent clinical reports, experimental research, and review articles.

Results

This review highlights our understanding of the CPI mechanism of action on cellular and molecular levels. Authors also discuss how reactivation of T cell responses through the inhibition of CTLA4, PD-1, and PD-L1 is utilized for tumor inhibition in cancer immunotherapy.

Conclusion

Mechanisms of PD-1 and CTLA4 blockade and normal biological functions of these molecules are highly complex and require additional studies that will be critical for developing new approaches to dissociate the benefits of checkpoint blockade from off-target effects of the immune reactivation that lead to immune-related adverse events.

Introduction

Cancer immunotherapy has revolutionized the field of medical oncology and significantly prolonged the survival of patients with life-threatening cancers. Prior to immunotherapy, most anti-cancer therapies were designed to target the tumor with agents that are toxic to rapidly dividing cells. Immunotherapy has shifted the paradigm in our approach to cancer treatment and moved the focus from cancer to the host, with latest therapies designed to enhance the endogenous immune response against tumor. In this review, we focus on the current understanding of checkpoint inhibitor (CPI) immunotherapy in cancer treatment. We believe this review would be of interest to the clinical and scientific audience of the Annals of Allergy, Asthma and Immunology, as cancer immunotherapies are becoming a standard of care for a variety of malignancies with a large number of patients now treated with CPI. Main side effects induced by CPI are broad-spectrum immune related adverse events (irAEs), with sometimes severe sequelae for the patient. As the number of patients treated with CPI is rising, allergists are increasingly confronted with patients with irAEs, which require new skills to successfully diagnose and manage. This review details the immunological mechanism of action of CPI antibodies and highlights potential links to the irAEs seen in response to CPI immunotherapy.

Role of CPI in the initiation and propagation of antigen-specific immune response

The two-signal model for T cell activation has been proposed as a model for the activation of naïve T cells. According to this model, T cells require two signals to become fully activated. The first signal, which gives specificity to the immune response, is provided by the interaction between the antigen and the T cell receptor (TCR). The antigen is presented as a peptide−major histocompatibility (MHC) complex by antigen presenting cell (APC) and is recognized by the TCR. This process typically occurs in lymph nodes. The second antigen-independent costimulatory signal is delivered to T cells by the APC to promote T cell clonal expansion, cytokine secretion, and effector function. This costimulatory signal is provided upon interaction of CD28, which is constitutively expressed on the cell surface of naïve CD4+ and CD8+ T cells, with CD80/CD86 molecules expressed by APCs.^1,2 Cytotoxic T lymphocyte antigen 4 (CTLA4; CD152) is an inhibitory receptor that suppresses the initial stages of T cell activation.^3,4 It competes with the CD28 receptor for the interaction with CD80 or CD86, thereby blocking T cell activation (Fig 1).

Figure 1.

Schematic representation of mechanisms of action of CTLA4 and PD-1-mediated inhibition of T cell activation.

The progression of T-cell activation, their inhibition by normal regulatory mechanisms in the presence of CTLA4 (A) and PD-1 (B), and release of these negative regulations by anti-CTLA4, anti-PD-1, and anti-PD-L1 antibodies during cancer immunotherapy are outlined.

Programmed cell death protein 1 receptor (PD-1, also known as PDCD1, CD279) is an inhibitory receptor expressed by the antigen-stimulated effector T cells.⁵ PD-1 has two known ligands, programmed cell death 1 ligand 1 and 2 (PD-L1; CD274 and PD-L2; CD273), which are expressed by myeloid cells or other non-lymphoid cells.^6,7 After interaction of PD-1 receptor with its ligands, PD-1 signaling inhibits T cell proliferation, cytokine release, cytotoxicity, and promotes T cell apoptosis (Fig 1).

Thus, normally, both CTLA4 and PD-1 provide immune checkpoints, and, therefore, inhibit T cell responses and protect the body from possibly damaging immune responses. CTLA4 regulates the establishment of de novo immune responses, while the PD-1 pathway exerts its major influence on ongoing (effector) immune responses. As the biological effects of CTLA4 and PD-1 are utilized at distinct body sites and during a specific time period in the T cell lifespan,⁸ these receptors functionally complement each other and regulate the balance between T cell immune tolerance to autoantigens and propagation of responses to foreign antigens, including infectious agents and tumors.

Tumors subvert CPI-regulated T cell response against tumor antigens

Human cancers carry somatic gene mutations and epigenetically altered genes, the products of which can be recognized by the immune system as foreign antigens. Although an endogenous immune response to cancer is observed in cancer patients, this response is not always efficient, because tumors induce tolerance among tumor-specific T cells and hijack the immune checkpoint system by expressing ligands that bind inhibitory receptors on T cells. This results in dampening T cell functions within the tumor microenvironment and allows tumors to evade the immune response.⁹

The checkpoint blockade (a blockade of immune checkpoint inhibitory pathways activated by cancer cells) has been developed as one of the approaches in cancer immunotherapy to reactivate anti-tumor immune response.⁹ For example, during the priming phase, naïve T cells in the lymphoid organs become exposed to tumor-specific antigens, resulting in the differentiation of naïve T cells into effector T cells (e.g. T regulatory cells [Treg], cytotoxic T cells, and helper T cells). This represents the initial step of an adaptive reaction against tumor cells, which is supported by the co-stimulatory effect of the CD28 receptor with CD80/86. The effect of CD28 is restrained in the presence of the CTLA4 receptor, which has a much higher affinity for the CD80/86 ligands. The rationale for using anti-CTLA4 antibodies in cancer immunotherapy is to unleash pre-existing anticancer T cell responses and possibly trigger new ones by inhibiting CTLA4 restraints on the formation of de novo immune responses (Fig 1).

In the effector phase, cytotoxic T cells in the tumor microenvironment eliminate tumor cells. This reaction is diminished by the interactions between the PD-1 receptor expressed by T cells and PD-L1, or, to a lesser degree, PD-L2, ligands expressed on the surface of tumor cells and host myeloid cells (i. e. macrophages) in the tumor microenvironment. Particularly, the interaction between tumor PD-L1 and PD-1 expressed by activated effector T cells results in blockage of production and secretion of cytotoxic mediators required for tumor killing. Antagonism of PD-1 or PD-L1 immune checkpoints by antibodies that target these receptors aims to maintain T-cell effector function and re-activate T-cell response against tumor cells (Fig 1).

CTLA4 and PD-1 have now been successfully targeted by multiple FDA approved drugs for cancer treatment (Table 1). James P. Allison and Tasuku Honjo were awarded the 2018 Nobel Prize in Physiology or Medicine for their pioneering work in this area.

Table 1.

FDA-approved immune checkpoint blocking antibodies

Target	Antibody Drug	Trade Name (Manufacturer)	Tumor Type (FDA Approval Year)
PD-1	Nivolumab (IgG4)	Opdivo (Bristol-Myers Squibb)	Melanoma (2014)55–57 Non-small-cell lung cancer (2015)89 Hodgkin lymphoma (2016)90 Head and neck squamous cell carcinoma (2016)91 Urothelial carcinoma (2017)92 Hepatocellular carcinoma (2017)93
	Pembrolizumab (IgG4)	Keytruda (Merck)	Melanoma (2014)54 Non-small-cell lung cancer (2015)94,95 Head and neck squamous cell carcinoma (2016)96 Hodgkin lymphoma (2017)97 Urothelial carcinoma (2017)98 Gastric and gastroesophageal carcinoma (2017)99
	Cemiplimab (IgG4)	Libtayo (Regeneron)	Cutaneous squamous cell carcinoma (2018)100
PD-L1	Atezolizumab (IgG1)	Tecentriq (Genentech)	Urothelial carcinoma (2016)101 Non-small-cell lung cancer (2016)102 Triple-negative breast cancer (2018)103 Small-cell lung cancer (2018)104
	Durvalumab (IgG1)	Imfinzi (AstraZeneca)	Urothelial carcinoma (2017)105 Non-small-cell lung cancer (2018)106
	Avelumab (IgG1)	Bavencio (EMD Serono)	Merkel cell carcinoma (2017)107 Urothelial carcinoma (2017)108
CTLA4	Ipilimumab (IgG1)	Yervoy (Bristol-Myers Squibb)	Melanoma (2011)29,30

CTLA4 blockade in cancer

CTLA4, an immunoglobulin superfamily member, is an immune checkpoint receptor expressed by T cells. It has structural and biochemical similarities to CD28, a T cell co-stimulatory receptor.^3,4 CTLA4 is expressed at a low level by naïve T cells and is induced following antigen stimulation. CD4+CD25+ Treg cells express CTLA4 constitutively. Due to structural similarities of the extracellular domains, both CTLA4 and CD28 interact with the same ligands expressed by APCs (CD80 and CD86). CTLA4 has a greater affinity and avidity for CD80/CD86 than CD28,^10–12 but has an opposite immunoregulatory function as it inhibits T cell activation and proliferation (Fig 1).^13–15 CTLA4 is essential for immune tolerance, as illustrated by CTLA4-deficient mice developing a T cell mediated lymphoproliferative autoimmune disease.¹⁶

To date, several mechanisms have been established for CTLA4-mediated regulation of T cell activation. These include direct antagonism of CD28, competition for co-stimulatory ligands, impediment of the immunological synapse formation and recruitment of signaling inhibitors.¹⁷

At the molecular level, the cytoplasmic tail of CTLA4 recruits phosphatases, mainly protein phosphatase 2A (PP2A). PP2A inhibits the activation of AKT serine/threonine kinase 1 (AKT1) and possibly extracellular signal regulated kinase (ERK).¹³ The recruitment of SH2 domain - containing tyrosine phosphatase 2 (SHP2) by the ligand-bound CTLA4 has been also suggested. SHP2 inhibits phosphorylation of CD3 subunits of the T cell receptor (TCR) and adaptor proteins, including the linker of activated T cells (LAT).^14,15 Downstream, these inhibitory signals result in reduced activation of transcription factors, such as activator protein 1 (AP1), nuclear factor κB (NFκB) and nuclear factor of activated T cells (NFAT), and reprogramming of T cells towards anergy (Fig 2).^21,22

Figure 2. Intracellular signaling by CTLA4 and PD-1.

CTLA4 engagement by its ligands, CD80 and CD86, activates the serine/threonine phosphatase PP2A, which directly inhibits the TCR/CD28-mediated activation of AKT.

PD-1 ligation by PD-L1 or PD-L2 leads to phosphorylation of tyrosine motifs in the PD-1 cytoplasmic domain, which results in recruitment of the tyrosine phosphatases SHP1 and SHP2, and inhibition of PI3K activity. PD-1 ligation also inhibits PLCγ1 and downstream Ras-MEK-ERK signaling. In contrast to PD-1, CTLA-4 has limited effects on Ras-MEK-ERK and PLCγ1 signaling inhibition.

CTLA4 expression on Treg is essential for both direct and indirect immunosuppressive activity of these cells.²³ CTLA4+ Treg cells control cytokine release and proliferation of effector T cells.²⁴ Treg cells also prime dendritic cells (DCs) to induce anergy in conventional T cells. It has been shown that CTLA4 on Treg cells can interact with CD80/CD86 ligands on APCs, resulting in internalization and degradation of these ligands and lack of co-stimulatory signals for conventional T cells.²⁵

Recognition of CTLA4 as a negative regulator of T cell activation suggested that its blockade could potentiate T cell responses against tumors.²⁶ Neutralizing CTLA4 antibodies were shown to enhance anti-tumoral immunity in mice against colon carcinoma and fibrosarcoma.²⁷ Animals treated with anti-CTLA4 that rejected tumors rapidly eliminated tumor cells that were introduced with the second challenge, providing evidence that blocking of CTLA4 induces long lasting immunological memory.²⁷

Monoclonal antibodies targeting CTLA4 proved effective in clinical trials of melanoma.²⁶ Ipilimumab, a humanized IgG1κ anti-CTLA4 monoclonal antibody, received FDA approval in 2011 for non-resectable stage III/IV melanoma, as it induced potent tumor necrosis²⁸ and provided a median 10.1 months short term survival benefit.²⁹ Long term survival data accumulated from multiple clinical trials demonstrated that 22% of ipilumab-treated patients with advanced melanoma lived 3 years or longer as compared to control standard of care treated patients³⁰ and provided evidence of persistent immunity against tumor following the CTLA4 blockade.³⁰ Despite the success of anti-CTLA4 inhibition for the treatment of melanoma, it was not found to be effective as a single agent in clinical trials for other cancers, including renal cell carcinoma, non-small cell lung cancer, small cell lung cancer and prostate cancer.⁹

Mechanisms of CTLA4-mediated tumor regression are attributed to the enhancement of effector T cell responses against tumor-associated neoantigens.^31,32 Anti-CTLA4 therapy has been suggested to deplete intra-tumoral Treg cells and therefore shift the immunoregulatory balance of tumor microenvironment away from immunosuppression.³³ Relative contributions of CTLA4 expression on T effector cells and Treg cells are currently under investigation and likely are dependent on the ratio of effector T cells to Tregs infiltrating the tumor.²⁶ Specific blocking of CTLA4 in both cell populations was shown to synergistically increase tumor regression.³³

In a clinical trial with anti-CTLA-4 (ipilimumab) administered to patients with localized bladder cancer prior to radical cystectomy, gene array data demonstrated that patients who responded to the therapy had significantly increased numbers of T cells expressing inducible co-stimulator (ICOS), a receptor closely related to the extended CD28/CTLA-4 family, in tumor tissues and blood.³⁴ The role of ICOS+CD4 T cells in the therapeutic effect of CTLA-4 blockade was confirmed in animal studies, where anti-CTLA-4 treatment resulted in tumor rejection in 80–90% of wild type mice, while in mice that were deficient for ICOS or its ligand, the efficacy was less than 50%.³⁵ Furthermore, an agonistic signal through ICOS increased the efficacy of CTLA4 blockade.³² Increased frequency of ICOS+CD4 T cells is now considered a pharmacodynamic biomarker of anti-CTLA-4 treatment.³⁷

The primary mechanism of action of CTLA4 blockade for tumor inhibition is the direct blockade of CTLA4 competition for CD80 and CD86 co-stimulatory ligands, which allows unrestrained CD28-mediated T cell activation. Crystallographic structural analysis of the ipilimumab:CTLA4 complex revealed that ipilimumab’s binding epitope overlaps with the CD80/CD86 interaction domain, indicating that steric inhibition of CD80/86 interactions underlies the primary mechanism of action of ipilumab.³⁸

Tumor cells do not express CD80/CD86 ligands, therefore it is presumed anti-CTLA4 inhibition occurs in tumor-draining lymph nodes where tumor antigens are presented by APCs to tumor-reactive T cells. Another possibility is a direct regulation of APCs within the tumor microenvironment, as these cells also present tumor antigens to activate competent T cells. CTLA4 blockade leads to the expansion of tumor neoantigen-specific CD8 T cells within the tumor microenvironment.³⁹ In addition to CTLA4 blockade-induced tumor rejection, depletion of Treg cells has also been identified as a contributing factor in anti-CTLA4 therapy in murine tumor models.⁴⁰ It was determined that blockade of CTLA4 in both the effector T cells and Tregs is required for effective tumor rejection.³³ However, the relative contribution of the effector function enhancement vs. Treg depletion to the efficacy of ipilimumab in humans remains unclear. Lastly, modulation of the TCR repertoire may contribute to the therapeutic effects of CTLA4 blockade against tumors.^41,42 Upon CTLA4 blockade, immunogenicity of low-affinity antigens that do not normally trigger effective T cell response may increase, while the activity of high-affinity tumor-reactive clones would also be boosted, resulting in more robust activation of tumor-reactive T cells, with an overall wider antigenic repertoire.

PD-1/PD-L1 biological function and its role in cancer immunotherapy

PD-1 is another immune checkpoint that limits the responses of activated T cells (Fig 1).⁵ PD-1, like CTLA4, has two ligands (PD-L1 and PD-L2) that, in contrast to CTLA4 ligands, are expressed on many different cell types. The function of PD-1 is distinct from CTLA4 in that PD-1 does not interfere with co-stimulation, but rather directly inhibits TCR⁴³ signaling.

PD-L1 is found on both hematopoietic cells (T cells, B cells, DCs, and macrophages) and non-hematopoietic cells (vascular and stromal endothelial cells, pancreatic islet cells, keratinocytes).^6,7 PD-L2 is expressed by DCs, macrophages, and B cells.^6,7 All T cells express PD-1 during activation. PD-1 is also expressed by subsets of Treg cells, T follicular helper (Tfh) cells, T follicular regulatory (Tfr) cells, memory T cells, and several other cell types, including B cells, natural killer (NK) cells, some myeloid cells and innate lymphoid cells.^6,7 It is also found on exhausted T cells during chronic infections and in cancer.

Much of our understanding of PD-1 signaling comes from studies of acutely activated T cells.⁴⁴ Upon interaction with the ligand, two tyrosine motifs in the PD-1 intracellular domain become phosphorylated, leading to binding of tyrosine phosphatases, such as SHP1 and SHP2. These phosphatases dephosphorylate kinases and inhibit signals that occur through the TCR and CD28, affecting downstream signaling pathways, including phosphoinositide 3-kinase (PI3K)-AKT, RAS, extracellular-signal-regulated kinase (ERK), and phospholipase C gamma (PLCγ) (Fig 2).^44–47 Functionally, this results in decreased T cell activation, proliferation, survival, cytokine production and altered metabolism.

Intracellular tails of PD-L1 and PD-L2 do not contain canonical signaling motifs. Some evidence, however, suggests that ligation of PD-1 ligands can also provide a signal for a PD-1 ligand-expressing cell.⁴⁸

There is increasing evidence for a functional link between PD-1 signaling and metabolic activity in T cells. PD-1 signaling modulates metabolic reprogramming during T cell activation, inhibiting the upregulation of glucose and glutamine metabolism that is driven by TCR and CD28 signaling.^45,49

The role of PD-1 and its ligands in the negative regulation of T cell activation was first illustrated in PD-1 (Pdcd1) deficient mice that develop autoimmunity: T cell-mediated lupus-like glomerulonephritis and arthritis was described in aged C57BL/6 PD-1 deficient mice;⁵⁰ BALB/c PD-1 deficient mice exhibit signs of cardiac inflammation;⁵⁰ non-obese diabetic Pdcd1^–/– mice develop accelerated type 1 diabetes.⁵¹ The function of PD-1 is distinct from CTLA4, as CTLA4 exerts its regulatory effects predominantly within lymphoid organs, while PD-1 controls T cell activation locally within peripheral tissues.² PD-1 acts later in the course of T cell activation and fate determination, and plays a unique role in maintaining T cell immune tolerance to autoantigens.

Following preclinical success, the efficacy of monoclonal antibodies that target the PD-1 axis was shown in clinical trials.⁵³ In 2014, humanized and fully human anti-PD-1 monoclonal antibodies pembrolizumab and nivolumab (both IgG4) became the first FDA approved PD-1-targeted therapeutics for refractory and unresectable melanoma.^41,54–56 In patients with melanoma, pembrolizumab showed a 6-month longer progression-free survival compared with ipilimumab and showed an overall survival benefit.⁵⁴ Clinical trials of nivolumab in melanoma demonstrated 72.9% overall survival in the first year in the immunotherapy treated group, compared to 42.1% overall survival in the control group of patients on chemotherapy (dacarbazine).⁵⁷ Additional successful clinical trials expanded the use of pembrolizumab and nivolumab to other types of cancer (see Table 1).⁹ Similar to PD-1, the blockade of PD-L1 has also been effective in many forms of cancer (Table 1).⁹

Despite active investigation, precise molecular and cellular events that mediate enhancement of anti-tumor immunity by PD-1 blockade are not fully understood. Blockade of the PD-1 signaling axis prevents PD-1-mediated attenuation of TCR signaling and restores activity of exhausted CD8 cells (Fig 1). Clinical evidence supports a model in which the blockade of PD-1 signaling is the most effective if the tumor antigen specific T cells were previously suppressed through PD-1 engagement by PD-L1 and PD-L2.^58–59 Although PD-1 blockade primarily leads to the expansion of CD8 T cells, CD4 T cells are additionally required for effective responses.⁶⁰ It has also been determined that anti-PD-1 therapy enhances neoantigen-specific T cell responses.⁶¹ PD-1 blockade induces the recovery of dysfunctional PD-1+ CD8+ T cells and enhances PD-1+ Treg cell-mediated immunosuppression. It has recently been determined that reactivation of effector PD-1+ CD8+ T cells, rather than PD-1+ Treg cells by PD-1 blockade is necessary for tumor regression. The frequency of PD-1⁺CD8⁺ T cells relative to PD-1+ Treg cells in the tumor microenvironment can predict clinical efficacy of PD-1 blockade therapies.⁶²

The inhibition of PD-L1 is thought to largely phenocopy the effect of PD-1 blockade. Activated T cells control PD-L1 expression by tumors, as PD-L1 is induced by Th1 cytokines (IFNγ), whereas PD-L2 is induced by Th2 cytokines.⁶³ However, new evidence suggests that anti-PD-1 and anti-PD-L1 therapies may not be mechanistically equivalent. For example, it has been demonstrated that PD-L1 inhibition can be partially attributed to antibody dependent cytotoxicity, with anti-PD-L1 antibody therapy inducing tumor regression in murine tumor models.⁶⁴ In addition, PD-L1 has been shown to interact with CD80/CD86, leading to the inhibition of T cell activity. This suggests a possibility for PD-1 independent control of T cells by PD-L1.⁶⁵

Profiling of immunotherapy response has determined that PD-L1 (and occasionally PD-L2) can be expressed on tumor and immune cells in the tumor microenvironment, the expression of PD-L1 by tumors was initially considered a selection criterion for patients to be treated with antibodies targeting the PD-1/PD-L1 pathway. Phase 1 trials with anti-PD-1 therapy (nivolumab) reported that PD-L1 expression on tumor cells prior to treatment, as detected by immunostaining, may serve as a predictive marker to indicate which patients would benefit from the treatment.⁶⁶ Patients with PD-L1-positive tumors (≥5% staining for PD-L1 on tumor cells) had an objective response rate of 36%, whereas patients with PD-L1-negative tumors did not show any objective clinical responses. In subsequent trials, however, some patients with PD-L1 negative tumors also demonstrated clinical responses to anti-PD-1 and anti-PD-L1 treatments with either tumor regression or stabilization of the disease.^57,58,67 Based on these observations, it was concluded that PD-L1 expression in tumor tissues cannot be used as a predictive biomarker for patient selection or exclusion for treatment with anti-PD-1 or anti-PD-L1 antibodies. However, it was noted that the presence of both intratumoral T cells and tumor PD-L1 expression is more beneficial for positive clinical outcomes of immunotherapy.^59,68 It was also noted that expression of other checkpoint receptors and their ligands in the tumor tissue could be important, as T cells that co-express PD-1 together with other inhibitory molecules, such as LAG-3 or Tim-3 may be more hyporesponsive than those expressing PD-1 alone, indicating potential benefits of controlling multiple checkpoints.^69,70

Genetic analyses of melanoma tumors revealed that higher numbers of mutations (“mutational load”) and neoantigens that can be recognized by T cells as a result of these mutations correlated with clinical responses to anti-CTLA-4 therapy.^31,71 Similarly, higher numbers of mutations with subsequent increase in numbers of neoantigens were found to correlate with clinical responses in patients with non-small cell lung cancer who received treatment with anti-PD-1 (pembrolizumab).⁷²

A clinical trial with anti-CTLA-4 (ipilimumab) in combination with anti-PD-1 (nivolumab) demonstrated enhanced efficiency, with tumor regression in about 50% of treated patients with advanced melanoma, most with tumor regression of 80% or more.⁷³ It was proposed that the enhanced efficiency was due to the fact that anti-CTLA-4 drives T cells into tumors, resulting in increased T cell infiltration and a concurrent activation of IFNγ production. This, in turn, can induce expression of PD-L1 in the tumor microenvironment, with subsequent inhibition of anti-tumor T cell responses, but also increases chances to benefit from anti-PD-1 and anti-PD-L1 therapies. Therefore, combination treatment with anti-CTLA-4 plus anti-PD-1 or anti-PD-L1 enables an immunogenic tumor microenvironment with subsequent clinical benefit for patients regardless of whether their tumor tissues have infiltrating T cells or express PD-L1 prior to treatment.

Combination studies with conventional anti-cancer agents (chemotherapy, radiation therapy) and immune checkpoint therapies demonstrated that this approach creates an “immunogenic” tumor microenvironment with subsequent clinical benefit for the patient. A number of experimental treatments combining CPIs with cancer vaccines, or new CPI therapies, or agonistic immune stimulatory pathways demonstrated clinical benefits in cancer treatment. Among these new target molecules are LAG3, TIM3, TIGIT, VISTA, and ICOS from the immunoglobulin superfamily, and OX40, GITR, 4–1BB, CD40, and CD27 from the tumor necrosis factor receptor superfamily. To date, our understanding of the biological roles of these molecules remains not fully understood and, in many cases, is outpaced by clinical investigations (discussed in references 74 and 75).

irAEs during the course of CPI immunotherapy

CPI immunotherapy, despite its obvious benefits for cancer treatment, is often accompanied by irAEs, which are believed to be the off-target effects of the activated immune response due to the elimination of immune checkpoints.⁷⁶ The frequency of irAEs is dependent on the types of CPI antibodies used, exposure time, and patients intrinsic risk factors. The timing of irAE appearance is often dictated by the affected organ systems.⁷⁷

IRAEs are estimated to occur in 15–90% of patients subjected to immunotherapy.⁷⁷ It has been determined that patients receiving anti-PD-1 or anti-PD-L1 antibodies have a lower incidence of any grade irAEs, than those receiving anti-CTLA4 antibodies, and patients receiving a combination of both classes of agents have the highest incidence of irAEs and the highest incidence of more severe irAEs (the incidence of grade ≥3 irAEs is 6%, 24%, and 55% of patients receiving anti-PD-1 or anti-PD-L1 antibodies, anti-CTLA4 antibodies, or a combination of anti-CTLA4 and anti-PD1 antibodies).⁷⁶ It was also noted that irAEs in patients receiving combination CPI inhibitors have an earlier onset than the same irAEs in those receiving monotherapies.⁷⁸ The lack of specificity in T cell expansion, coupled with the fundamental importance of CTLA4 as an immune checkpoint in the formation of de novo immune response, could account for the significant irAEs observed in patients treated with anti-CTLA4. The favorable toxicity profile of the PD1- and PD-L1 blocking agents hint at the benefits of specifically targeting the properties of cancer that inhibit the immune response rather than non-specific activation of the immune system with anti-CTLA4 blockade.

Due to the diversity of the T cells and their ability of these cells to infiltrate most organs, CPI can cause a wide range of irAEs and these can affect virtually any organ. Prominent adverse events have been registered at various mucosal sites,^79,80 including some allergic manifestations^77,80,81 and autoimmune specific reactions.⁸² This is discussed in detail in the articles by Wang et al.⁸³ and Lyubchenko et al.⁸⁴ in this issue of Annals of Allergy, Asthma and Immunology. The most frequently affected organs and the most common specific irAEs include cutaneous irAEs (rash, pruritis, bullous, lichenoid, eczema-like and psoriasis-like manifestations), lower digestive tract irAEs (colitis), pulmonary irAEs (pneumonitis), endocrine irAEs (hypophysitis, thyroiditis), hepatic irAEs (hepatitis), cardiac irAEs (myocarditis), and other rare irAEs (neurological, renal, ocular).^76,77 Fatal irAEs are rare (estimated incidence between 0.3% and 1.3%) and are mainly attributed to severe cases of colitis, pneumonitis, hepatitis and myocarditis.⁷⁶ The mechanisms behind the development of tissue specific types of irAEs are not fully developed, but are suggested to involve reactivation of self-reactive T and B cell clones, inhibition of T regulatory cell function, interaction of CPI antibodies with the CPI receptors expressed by other tissues, immunogenisity of the CPI antibodies, potential antibody dependent cytotoxicity, and antibody dependent complement activation reactions.⁷⁶

We would like to emphasize that majority of CPI-related irAEs are managed by corticosteroids, which remain the cornerstone for the irAEs treatment. Personalized approaches, including other immunosuppressive medications, monoclonal antibodies are now discussed and started to be implemented for management of specific irAE manifestations in patients on CPI immunotherapy.⁸¹ For the detailed discussion of irAE types that develop in response to CPI and current management approaches, we would like to refer the readers to several excellent reviews that have been written over the last year.^76,77,81 For the discussion about cutaneous irAEs on CPI immunotherapy, please refer to the review by Wang et al.⁸³ in this issue of Annals of Allergy, Asthma and Immunology.

Conclusion

The dynamic interplay between the immune system and cancer is complex. As both the tumor and the immune infiltrate influence responses to therapy, both need to be considered when developing combination therapies.

CPI therapy provided a paradigm-shifting approach for cancer therapy. However, despite all of the advancements, clinical findings underscore the need for a better mechanistic understanding of checkpoint inhibition pathways; as for some patients, modulation of these pathways leads to significant clinical benefit, while in other cases the effect is only temporary, partial, or absent. Importantly, acquired resistance to CPIs has been described,⁸⁵ including the contribution of microbiome and microbial metabolites to CPI immune responses.⁸⁶ IRAEs are estimated to occur in 15–90% of patients subjected to immunotherapy.⁷⁷

It is currently unclear whether IRAEs arising after immunotherapy in patients with cancer are caused by a pre-existing autoimmune response that is worsened by the checkpoint blockade, or whether IRAEs are caused de novo by priming of autoreactive lymphocytes in these patients. Importantly, many tumor antigens are likely to be similar to self-antigens. CPIs may work, in part, by boosting autoreactive T cells that are specific for self-antigens in the tumor.⁸⁷ Determining the immunological origins of IRAEs will guide a safer use of CPIs in cancer immunotherapy.

We would like to emphasize that the mechanisms of action of PD-1 and CTLA4 blockade and normal biological functions of these molecules are highly complex and are not fully understood.⁸⁸ It is likely that subtle nuances in timing, kinetics, target cell type, antigen availability, and anatomic location may have profound impacts on the final outcome of the inhibition of these molecules.

Another critical question is the degree to which the manifestation of IRAEs is functionally and mechanistically associated with therapeutic efficacy of CPIs. If distinct mechanisms are responsible for these biological responses, it is possible that mechanisms underlying the efficacy and IRAEs could engage separately. It would be important to find out whether it is possible to uncouple the anti-tumor activity of immune checkpoints from the function of these checkpoints in maintaining self-tolerance to limit IRAEs.

Additionally, of central importance to the mechanisms of action of immune checkpoint blockade therapy is understanding what properties define the antigens that are recognized and mediate tumor rejection.

We would like to point out that there is a need to understand beyond conventional CD4+ and CD8+ T cells, what are the functions of PD-1 on Treg cells, B cells, myeloid cells, innate lymphoid cells and NK cells. Additional work is also needed to understand how PD-1 controls these cells, as systemic modulation of the PD-1 pathway likely broadly impacts all immune cell subsets.

In this review, we have highlighted known mechanisms of anti-CTLA4 and anti-PD-1 immune CPIs. Mechanistic insights into the biological functions of these molecules are important for the development of new approaches and improvement of immunotherapeutic strategies in cancer.

Key messages:

• In recent years T cell-based immunotherapies have moved to the cutting edge cancer treatment. These include immune checkpoint blockade, an approach designed to activate or reinvigorate anti-tumor T cell responses.

• Cytotoxic T lymphocyte antigen 4 (CTLA4) and programmed cell death protein 1 (PD-1) T cell receptors function as “brakes” for the adaptive immune response and immune checkpoints for effector T cells.

• Humanized monoclonal antibodies against CTLA4, PD-1, and its ligand PD-L1 are now successfully used as monotherapy or in combination with other treatments for a wide variety of cancers.

• Mechanisms of CTLA4-mediated tumor regression are attributed to the enhancement of anti-tumor effector T cell responses through elimination of antagonistic competition with CD28 for co-stimulatory ligands, and to the inhibition of CTLA+ T regulatory cell mediated immunosuppression within the tumor.

• PD-1 controls T cell activation locally within peripheral tissues. Blockade of the PD-1 signaling prevents PD-1-mediated attenuation of the T cell receptor signaling and restores anti-tumor activity of “exhausted” CD8+ T cells and/or prevents the exhaustion of newly generated anti-tumor specific T cells. The inhibition of PD-L1 largely phenocopies the effect of PD-1 blockade.

• Despite the clinical success, mechanisms of PD-1 and CTLA4 blockade and their normal biological functions are not fully understood and require further investigation as the timing, kinetics, target cell type, antigen availability, and anatomic location have significant impact on final outcomes of inhibition of these molecules in cancer immunotherapy.

Abbreviations:

AKT1

AKT serine/threonine kinase 1

AP1

activator protein 1

APC

antigen presenting cells

CPI

checkpoint inhibitor

CTLA4

cytotoxic T lymphocyte antigen 4

DC

dendritic cells

ERK

extracellular signal regulated kinase

ICOS

inducible co-stimulator

IFNγ

interferon gamma

IRAE

immune related adverse event

LAT

linker of activated T cells

NFAT

nuclear factor of activated T cells

NK cells

natural killer cells

NFkB

nuclear factor κB

PD-1

programmed cell death protein 1

PD-L1

programmed cell death protein 1 ligand 1

PD-L2

programmed cell death protein 1 ligand 2

PI3 kinase

phosphoinositide 3-kinase

PLCγ

phospholipase C gamma

PP2A

protein phosphatase 2A

SHP1

SH2 domain containing tyrosine phosphatase 1

SHP2

SH2 domain containing tyrosine phosphatase 2

TCR

T cell receptor

Treg

T regulatory cells

Tfh

T follicular helper cells

Tfr

T follicular regulatory cells