Immune interactions in pembrolizumab (PD-1 inhibitor) cancer therapy and cardiovascular complications

Abstract

The use of immunotherapies like pembrolizumab (PEM) is increasingly common for the management of numerous cancer types. The use of PEM to bolster T-cell response against tumor growth is well documented. However, the interactions PEM has on other immune cells to facilitate tumor regression and clearance is unknown and warrants further investigation. In this review, we present literature findings that have reported the interactions of PEM in stimulating innate and adaptive immune cells, which enhance cytotoxic phenotypes. This triggers secretion of cytokines and chemokines, which have both beneficial and detrimental effects. We also describe how this leads to the development of rare but underreported occurrence of PEM-induced immune-related cardiovascular complications that arise suddenly and progress rapidly to debilitating and fatal consequences. This review encourages further research and investigation of PEM-induced cardiovascular complications and other immune cell interactions in patients with cancer. As PEM therapy in treating cancer types is expanding, we expect that this review will inform health care professionals of diverse specializations of medicine like dermatology (melanoma skin cancers), ophthalmology (eye cancers), and pathology (hematological malignancies) about PEM-induced cardiac complications.

INTRODUCTION

Immunotherapy is an alternative strategy in the treatment of advanced-stage cancer. Advancements in immunotherapy have revolutionized care, as it is being used to manage cancers that were previously known to have poor diagnosis. Immunotherapy harnesses the immune system’s capability to eliminate tumors. The two main methods of immunotherapies used to combat cancer are to stimulate antitumoral immunity and/or to enhance specificity and targeting of cancer cells. Therapeutic approaches for this include cytokine therapy, adoptive cell therapy, cancer vaccines, and immune checkpoint inhibitor therapy. Cytokine therapy was the first immunotherapy approved in the United States in 1992 by the Food and Drug Administration (FDA), where interleukin-2 (IL-2) was used to treat renal cell carcinoma (1, 2). Cytokine administration of IL-2 and interferon alpha (IFN-α) has been used clinically to aid in T-cell growth and maturation (1, 2). Adoptive cell transfer therapies include tumor-infiltrating T cells (TILs) and chimeric antigen receptor T cells (CAR-T cells). TILs are isolated T cells from the patient, expanded in vitro to exhibit higher cytotoxicity and then transplanted intravenously (IV) to patients. Similarly, CAR-T cells are T cells collected from the patient, genetically modified with a receptor that enhances T-cell cytotoxicity and transplanted back. Both approaches involve the manipulation of T cells to exhibit a mature cytotoxic phenotype, with CAR-T cells having an added antigen-specific targeting feature established by genetically engineering the T-cell receptor (TCR) (3). Whereas, cancer vaccines involve administration of tumor antigens combined with immunological active cells or activating agents that reenergize immunity against cancer (4). Immunotherapies are delivered systemically to trigger the body’s natural immune response to cancer (5, 6). Immune checkpoint inhibitor therapies work by blocking immune-silencing mechanisms and invigorating immune cells that become anergic because of interactions with the immunosuppressive tumor microenvironment (7–9).

Among immune checkpoint inhibitor therapies on the market, pembrolizumab (anti-PD-1) is most widely used in the treatment of diverse cancer types (10). Pembrolizumab (PEM) also known as Keytruda, developed by Merk & Co., is a humanized monoclonal antibody, initially approved for the treatment of refractory unresectable or metastatic melanoma in 2014 by the FDA (11, 12). PEM was the first approved inhibitor for programmed cell death protein 1 (PD-1), which binds and prevents activation of PD-1 receptors located on T cells (11, 13). Similar therapeutic approaches are summarized in Table 1, target T-cell stimulation and inhibition. CTLA-4 (cytotoxic T lymphocyte-associated antigen 4), another T-cell inhibitory receptor, has been used in conjunction and/or independently with PD-1 inhibition. In addition, emerging T-cell receptor-targeting therapies are being pursued (Table 1). Prompted by the success of immunotherapies, T-cell receptor-targeting strategies, like PD-1 inhibition, are quickly gaining traction and show promise in preclinical trials (50, 51).

Table 1.

Immune regulatory receptor drug targets

Stimulatory Receptors	Effector Signaling Pathways Upregulated	Inhibitory Receptors	Effector Signaling Pathways Downregulated
CD28 (14–17)	PI3K-AKT, ERK, T-bet, Eomes, GATA3, AP-1, NFAT, NF-κB (18, 19)	CTLA4 (CD152) (14, 16, 17, 20)	PI3K-AKT, AP-1, NF-κB, NFAT, MAPK (21, 22)
ICOS (14–17)	JNK, PI3K-AKT-mTOR-NFAT (23, 24)	PD-1 (CD279) (14, 16, 17, 20)	MAPK, PI3K-AKT, AP-1, NFAT, NF-κB (21, 25)
HVEM (17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT, AP-1 (26, 27)	BTLA (17, 28)	Binds with CD160, reduced CD3 signaling, NF-κB, AKT, STAT3 (29–31)
CD27 (15–17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27, 32)	TIM-3 (14, 17, 20)	Binds to Galectin-9/Bat3, PI3K-AKT, NF-κB (33, 34)
4-1BB (CD137) (14–17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27, 35)	TIGIT (14, 17, 20)	ZAP70/Syk, ERK (36), PI3K (37)
OX-40 (CD134) (14–17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27, 38)	CD160 (17)	Binds with BTLA, reduced CD3 signaling (30, 31), p56, PI3K (26)
DR3 (17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27, 39)	LAG-3 (14, 17, 18)	STAT5, AKT (40)
GITR (14, 17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27)	LAIR1 (17)	JAK/STAT, MAPK (26)
CD30 (16, 17)	TRAF recruitment, NF-κB, MAPK, PI3K-AKT (27)	B7-1 (14, 17)	NF-κB (41)
SLAM (15, 17)	PI3K/Akt/mTOR, NF-κB, MAPK (42)	NRP1 (20)	MAPK, PI3K (43)
CD2 (17)	mTOR, AMPK, MAPK (44)	CCR4 (45)	MAPK, PLC, PI3K (46)
TIM-2 (17)	Largely still unknown	VISTA (47)	Largely still unknown
CD226 (14, 17)	PI3K-AKT, ERK, FOXO1 (48)
LFA-1 (CD11A/CD18) (15)	PI3K (49)

Immune stimulatory and inhibitory receptors are common drug targets for immunotherapy. Among these targets is PD-1, the target for pembrolizumab. Other drug-targeting therapies have been used in conjunction with pembrolizumab to rapidly reduce and eliminate tumor burden. AMPK, AMP-activated kinase; AP-1, activator protein 1; BAT3, B-associated transcript 3; BTLA, B- and T-lymphocyte attenuator; CCR4, C-C chemokine receptor type-4; CD, cluster of differentiation; CTLA-4, cytotoxic T-lymphocyte antigen-4; eomes, eomesodermin; ERK, extracellular signal-regulated kinase; FOXO1, forkhead box protein O1; GATA3, GATA binding protein 3; GITR, glucocorticoid-induced TNFR-related protein; HLA, human leukocyte antigen; HVEM, herpes virus entry mediator; ICOS, T-cell costimulator; JAK, janus kinase; LAG-3, lymphocyte activation gene 3; LAIR1, leukocyte-associated immunoglobulin receptor 1; LFA-1, leukocyte function-associated antigen 1; MAPK, mitogen-activated protein kinase; mTOR, mammalian target of rapamycin; NFAT, nuclear factor of activated T cells; NF-κB, nuclear factor κB; NRP1, neuropilin-1; PD-1, programmed cell death protein 1; PI3K-AKT, phosphatidylinositol 3-kinase-protein kinase B; PLC, phospholipase C; RAS, rat sarcoma virus; SLAM, signaling lymphocytic activation molecule; STAT, signal transducer and activators of transcription; syk, spleen tyrosine kinase; T-bet, T-box expressed in T cells; TIGIT, T-cell immunoglobulin and ITIM domain; TIM, T-cell immunoglobulin and mucin domain; TRAF, tumor necrosis factor receptor-associated factor; VISTA, V-domain Ig suppressor of T-cell activation; ZAP70, ζ-chain-associated protein kinase 70.

PD-1 is essential in the prevention of autoimmune diseases by maintaining immune tolerance, where healthy tissues are protected from strong immune reactions like inflammation (8, 52).

While the role of PD-1 in T cells is well established, recent studies have shown the presence of PD-1 on B cells, macrophages, and dendritic cells (DCs) (53). Little is known about the role of PD-1 in macrophages, DCs, and B cells, as expression is low (53–55). Understanding the role of PD-1 in other immune cells could provide insight on the overall consequences of targeted anti-PD-1 therapy (55).

Targeted anti-PD-1 therapy has been very efficacious in the treatment of many types of cancers including bladder, lung, skin, as well as various types of carcinomas and lymphomas; however, durable response is rare (56). Acquired resistance and immune-related adverse events (irAEs) including cardiovascular disease (CVD) complications from therapy have been observed (56–58). These CVD complications include cardiac fibrosis, myocarditis, and heart failure that are largely irreversible (51, 57).

Unfortunately, the incidence and pathophysiology of CVD complications from PEM treatment are unknown (57). Many preclinical studies have indicated PD-1, PD-L1, and PD-L2 have protective roles in CVD, and its blockade might lead to adverse cardiac effects (58). Animal models and clinical cases show newly acquired CVD complications from PEM therapy (57, 59). While these complications may be rare, they progress rapidly and are associated with high morbidity and mortality (51, 60). Better understanding of CVD-related irAEs is critical in optimizing care. This is especially important due to the anticipated rise in patients treated with PEM and continued expansion in the use of this drug for different cancer types and indications (61).

In this review, we summarize PEM-mediated antitumoral response provided by immune cells like T cells, macrophages, and DCs, and briefly discuss CVD complications that arise from PEM treatment.

PEMBROLIZUMAB STRUCTURE

PEM was grafted from mouse antihuman PD-1 antibody and a S228P mutated human IgG4 containing a proline replacing serine residue (62). PEM has many of the same structural characteristics of naturally occurring IgG4 antibodies found in the body (Fig. 1). All IgG antibodies include N-linked glycosylation site at asparagine (Asn) residue 297 that helps to stabilize the fragment crystallizable region (F_c) region (63). In PEM, both CH2 domains on the heavy chain portion of the antibody contain this glycan modification (Fig. 1). The core of this glycan moiety comprises N-acetylglucosamine, mannose, and fucose (7). This glycan residue is essential in IgG antibodies, as it is involved in regulating immune effector functions including the release of cytokines, activation of antibody-dependent phagocytosis, and complement-mediated cytotoxicity (7, 64). However, X-ray structures of PEM indicate glycans are buried and have limited reactivity (7).

Figure 1.

Pembrolizumab structure—pembrolizumab (Keytruda), a monoclonal humanized IgG4 antibody. CH, constant heavy; CL, constant light; Fab, antigen-binding fragment; F_c, crystallizable fragment; VH, variable heavy; VL, variable light.

A unique feature of PEM are the S228P mutations in the hinge region (7). PEM has a shorter hinge region than other antibodies due to a serine (S) to proline (P) exchange at position 228 (7). This S228P mutation located in the hinge region, stabilizes interchain disulfide bond formation between cysteine residues contributing to tight interactions in the upper hinge region (7). This conformational change leads to a shorter and more compact hinge region, causing steric constraints (7). This conformational constraint hinders the ability of the antibody to go through antigen-binding fragment (F_ab) arm exchange (FAE), a naturally occurring process in which there is an exchange of one heavy and light chain between IgG4 antibodies, resulting in bispecific antibodies (65, 66).

Alterations in the hinge region and Fc domain contributes to a lower affinity for effector immune cells and complement protein C1q (7). This is advantageous as it decreases the chance of rejection to PEM therapy (65).

The structural features of PEM preserve the antibody’s engagement potency and pharmacodynamic activity (62). The low affinity for F_c receptor restricts the formation of immune complexes that would activate effector mechanisms leading to the clearance of the PEM antibody (65, 67). PEM has been shown to have profound effects in reducing tumor burden; however, treatment administration is often interrupted because of the occurrence of adverse reactions (62). These insights on the structural features allow for the antitumor effect of PEM therapeutic antibodies (62). Further investigations of the structural features that contribute to PEM physiological activity can help to further optimize safety and efficacy of the treatment (62).

IMMUNITY IN THE TUMOR MICROENVIRONMENT

Cancer immunosurveillance is a general term used to describe the active role host immunity plays in protecting and preventing cancer development by actively participating in recognition and elimination of tumor cells (68). This idea has since expanded to include the concept of immunoediting. In immune surveillance, the immune system actively recognizes and prevents tumor growth, whereas immunoediting, a more recent idea, acknowledges that host immunity plays a significant architectural role in evoking tumor cell outgrowth and ascending virulence (68, 69). Collaborative efforts between tumor and surrounding stromal environment are critical in modulating cancer cell growth and differentiation (70). Phases of immune editing include elimination, equilibrium, and escape (68) (Fig. 2).

Figure 2.

Immunity in the tumor microenvironment-collaborative interactions between the tumor and immune cells lead to tumorigeneses. Regulatory MDSCs, Tregs, and M₂ macrophages are immune regulatory subsets involved in actively suppressing immunity allowing for tumor escape and metastasis. M₂, macrophages; MDSCs, myeloid-derived suppresser cells; NK, natural killer; Tregs, T-regulatory cells.

At the elimination stage, the immune system effectively removes tumor cells and offers adequate protection. This then progresses to the next stage of equilibrium, where tumor growth becomes chronic and persistent. This is where tumor outgrowth and immunity are at equal footing and results in a high degree of inflammation (68). Eventually, tumor cells escape and overwhelm the immune system by evading recognition and increasing resistance and survival, while actively participating in immunosuppression (69). Targeting these phases of immunoediting is how immunotherapies augment immunity to appropriately target tumor cells (68).

Tumor infiltration by immune cells adds a layer of complexity and is a critical component of tumor pathogenesis. These immune cells can be broadly classified into the innate and adaptive immune cells (70). Innate cells include mast cells, DCs, granulocytes, macrophages, and natural killer cells (9, 70–72). This type of immune response is immediate and contributes to a quick inflammatory reaction. The adaptive system, on the other hand, develops slowly and is more targeted (70). B and T cells make up the adaptive immune response. This targeted response is facilitated by unique and specific antigen-recognizing immunoglobulin molecules and immune cell receptors that recognize processed peptides displayed on major histocompatibility complex (MHC) presented by antigen-presenting cells (APCs) (70). B cells, DCs, and macrophages are APCs that enables adaptive cells to recognize antigens (73). The degree and type of immune infiltrates correlate metastatic potential and prognosis of cancer.

Metastasis is the result of tumor cells escaping immunosurveillance, where tumor cells acquire the necessary characteristics to circumvent immune detection, and results in failure of immunity to combat cancer. This failure can arise from the aberrant activation of regulatory pathways typically used in maintaining immune tolerance (74). Chronic-phase immune failure against cancer is an important problem that arises from the activation of immune inhibitory receptors resulting in tumors escaping immune detection. Among these inhibitory receptors, PD-1 plays a critical role in exhaustion of immune cells (75). Increased expression of the ligand for PD-1, PD-L1, is associated with increased infiltration of regulatory T cells (Tregs), myeloid-derived suppresser cells (MDSCs), and anti-inflammatory M₂ macrophages, which are immunosuppressive (72, 76, 77). PEM strategically blocks PD-1/PD-L1 interaction and prevents tumor escape (Fig. 3).

Figure 3.

Pembrolizumab mechanism of action–pembrolizumab blocks PD-1 and PD-L1/PD-L2 interaction preventing deactivation of T-stimulatory signals 1 (antigenic stimulation through TCR-MHC interaction) and 2 (costimulation) responsible for active antitumoral immunity. APC, antigen-presenting cell; CD3ζ chain, cluster of differentiation 3ζ-chain; ITIM, immunoreceptor tyrosine-based inhibitory motif; ITSM, immunoreceptor tyrosine-based switch motif; MHC, major histocompatibility complex; PD-1, programmed cell death protein 1; PD-L1/PD-L2, PD-1 ligand; SHP-1/SHP-2, SH2 domain-containing tyrosine phosphatase; TCR, T-cell receptor; ZAP-70, ζ-chain-associated protein kinase 70.

PEMBROLIZUMAB MECHANISM OF ACTION

PEM is a 146.3-kDa IgG4 monoclonal antibody that binds with high affinity (29pM) to immune receptor PD-1 (44) (Fig. 3). The binding of PEM to PD-1 prevents interaction with PD-L1/PD-L2 ligands inhibiting the activation of the downstream immune inhibitory pathways, leading to sustained potent antitumoral response (74). Activation of PD-1 (also known as CD279) contributes to a rapid de-escalation of T-cell immune response. Upon association with PD-L1 (CD274) or PD-L2 (CD273), T cells undergo downstream signaling pathways that promote an overall anti-inflammatory state. Activation of this pathway leads to decreased T-cell proliferation and survival (56). Tumor cells exploit this pathway by expressing high levels of PD-L1/PD-L2, leading to reduced T cell-mediated immune response and contributing to increased tumor aggression and poor prognosis (11, 78). PEM intervenes by binding to PD-1 receptors on T cells, which eliminates association with its ligand and averts T-cell anergy (51, 79).

PD-1 has two important regulatory phosphorylation motifs, immunoreceptor tyrosine-based inhibitory motif (ITIM) and immunoreceptor tyrosine-based switch motif (ITSM). Upon activation, PD-1 is phosphorylated at intracellular domains containing tyrosine residues that trigger the activation of SH2-domain containing tyrosine phosphatases (SHP-1 and SHP-2) (72, 80). SHP-1 and SHP-2 bind to regulatory motifs ITIM and ITSM. SHIP-1 and SHP-2 deregulate phosphatidyloinositol-3-kinase/protein kinase B pathway (PI3K/AKT), silencing stimulatory feedback triggered by antigen recognition (80, 81). PD-1 activation inhibits effector functions in immune cells (51). PD-1 is mostly expressed in T cells and present to a lesser extent in other immune cells including dendritic cells, B cells, natural killer (NK) cells, monocytes, and macrophages (72, 80). Although PD-1 is found in many immune cell types, therapeutic inhibition of this receptor is most potent in T-cell populations (82). Engagement of PD-1/PD-L1 in both adaptive and innate immune subsets have the universal effect of facilitating immune-dampening phenotypes and reducing immune cytotoxicity and survival (51, 75, 83, 84).

ANTI-PD-1 ANTIBODY THERAPY IN T CELLS

T-cell stimulation is achieved through two mandatory signaling pathways that include antigenic stimulation and costimulation from APCs (Fig. 3). Antigenic stimulation occurs between the T-cell receptor (TCR) and antigen presented on MHC receptors on APCs. This is responsible for specificity in immune response. The second mandatory signal is costimulation by APCs. This signal contributes to increased T-cell effector function like T-cell proliferation, survival, differentiation, and cytokine secretion.

When T cells receive the first antigenic stimulation, activated PD-1 recruits SHP-1/2 to the immunoreceptor tyrosine-based switch motif (ITSM) to inactivate TCR activation downstream mediators, cluster of differentiation 3ζ a chain (CD-3ζ), ζ-chain-associated protein kinase 70 (ZAP70), and PI3K/AKT (72, 85) (Fig. 3). PD-1 activation also leads to the downregulation of antiapoptotic mediator B-cell lymphoma-extra large (Bcl-xl) (72, 86). Furthermore, activation of PD-1 leads to the differentiation of T-effector cells into regulatory T cells (Tregs) (71, 87). A third optional stimulation mechanism can be achieved through cytokine signaling. Cytokine signaling can help overcome activation of inhibitory pathways including PD-1 activation (58, 80). For example, PD-1 engagement reduces extracellular receptor kinase (ERK) signaling that can be overcome by expanding the signal transducer and activator of transcription 5 (STAT5) activated by interleukin-2 (IL-2), IL-7, and IL-15 (80, 88). Moreover, anti-PD-1 antibody therapy abolishes PD-1 activation, and T-cell antigenic stimulation is preserved.

ANTI-PD-1 ANTIBODY THERAPY ON INNATE IMMUNE CELLS

PD-1 interactions with innate immune cells is indirect where binding interactions between anti-PD-1 antibodies and innate immune cells are slight; however, indirect interplay exists where overall immune regulation provides significant therapeutic efficacy (89). Woo et al. (90) reported that while blocking PD-1/PD-L1 interactions to promote an antitumoral response in wild-type mice was observed, this effect was abolished in stimulator of interferon genes (STING)-deficient mice. The STING pathway is an innate immune-sensing pathway, essential for eliciting immune response upon detection of tumor-derived DNA, used by tumor-infiltrating DCs and macrophages (6, 91, 92). This observation emphasizes the role of innate immune-sensing pathways used to mediate immune response against tumors, critical in the efficacy of anti-PD-1 therapy. Downstream effects of PD-1 activation are presumed to be ubiquitous among immune cells. It has been noted that the engagement of PD-1 leads to an inhibition of PI3K-AKT-NF-κB (phosphatidyloinositol-3-kinase-protein kinase B-nuclear factor-κB) signaling, resulting in downregulation of inflammatory cytokines like IL-12 and tumor necrosis factor α (TNF-α). In addition, previous studies show that PD-1 signaling in DCs and macrophages stimulated by T-cell receptors leads to the downregulation of several proinflammatory cytokines that include TNF-α, IL-6, and IL-12 (93). Consideration of both direct and indirect effects of anti-PD-1 therapy in the tumor microenvironment is important when evaluating the underlying mechanisms behind tumor rejection that occurs during chronic therapeutic interventions (94). Furthermore, this information can be used to predict the consequence of integrating PD-1 inhibition with innate cell-based vaccines that have gained significant interest (93).

ANTI-PD-1 ANTIBODY THERAPY IN MACROPHAGES

Macrophages are unique in that they can influence both the innate and adaptive immune response by acting as APCs, participating in tissue remodeling, and releasing antimicrobial agents. However, due the lack in antigen specificity, macrophages are innate immune cells, differing significantly from adaptive immune cells like T cells where PD-1 was first discovered (53). In macrophages, PD-1 expression is triggered by pattern recognition receptors (PRRs) that corresponds to the activation of transcription factor NF-κB (75, 95). Tumor-infiltrating macrophages are critical to immune response to tumor growth. Although PD-1 inhibition is highly effective in abrogating tumor growth, this effect can be abolished by the depletion of tumor-associated macrophage (77).

Broadly, macrophages are categorized into M₁ and M₂ subsets depending on phenotype (9). M₁ macrophages are polarized toward an inflammatory state and participate in antigen presentation (9). Whereas M₂ macrophages are immunosuppressive, anti-inflammatory in nature, and participate in wound healing (9, 53, 77). Gordon et al. (77) reported that there is a positive correlation between the population of tumor-associated macrophages (TAMs) expressing PD-1 and duration of cancer progression. They found that at advanced cancer stages, increased PD-1-expressing macrophages shift to an M₂ immunological profile, a shift favorable to tumor progression (77). Interactions between macrophages expressing PD-1 and tumors expressing PD-L1 cause macrophages to reduce phagocytosis capability and fail to respond to tumor growth (77).These cells also undergo changes in cytokine production, shifting from proinflammatory IL-6 to anti-inflammatory IL-10- and TGF-β-producing cells (76, 96). Treatment with anti-PD-1 revealed significant reduction in tumor size and abrogated PD-1-mediated immune inhibition of macrophages, rescuing effector function such as phagocytosis of tumor cells (77).

In another study, time-lapse intravital microscopy in a mouse model was done by Arlauckas and colleagues (89) to provide a further look into anti-PD-1 distribution and interactions in the tumor microenvironment. They found several interesting results. First, after the administration of the drug over time, there was an exchange of bound anti-PD-1 antibodies from T cells to macrophages. They found that the uptake by macrophages occurred because of the interactions between the crystallizable fragment (F_c) domain of the antibody and the F_c gamma receptor (F_cγR) on macrophages. Second, they found that if they blocked F_c and F_cγR interactions, they could prevent anti-PD-1 uptake by macrophages and regain therapeutic efficacy (89). Researchers noticed that macrophage uptake was enhanced when anti-PD-1 binds to T cells. This is because F_cγR binds more readily to immune complexes. They also found that once anti-PD-1 was removed from the T-cell surface, there was an absence of reattachment with a new antibody. They concluded that antibody transfer from T cells was occurring at a faster rate than antibody uptake by T cells in the tumor microenvironment (89). Therefore, intervention by targeting the F_cγR before anti-PD-1 therapy can enhance potency of PEM.

ANTI-PD-1 ANTIBODY THERAPY IN DENDRITIC CELLS

Dendritic cells (DCs) are potent antigen-presenting cells (APCs) critical in the conservation of immune balance (97). This balance is maintained through the regulation of populations of mature and immature DCs. Mature DCs are activated effector immune cells that carry antigens and stimulate the adaptive immune response. In contrast, immature DCs take up self-antigens and sustain peripheral tolerance. These cells have reduced levels of MHC II and costimulatory receptors and work to suppress effector immune reactions.

Due to the potency of DCs in actuating and aggravating immune reactions, there is strict regulation of DC life span that ensures appropriate levels of mature and immature DCs. Once DCs are activated, there is a negative feedback mechanism where DCs are eliminated from the immune population soon after antigen presenting occurs to limit activation of T cells. DC and T-cell interactions modulate overall DC homeostasis, directing the life span of DCs through induction of survival and apoptosis signaling pathways, essential in limiting unrestrained immune reactions (93, 97).

Work by Park and others (93) show that PD-1 receptor signaling in DCs contribute to DC apoptosis. PD-1 signaling can occur between DCs or through DC and T-cell interactions. This signaling interaction is specific to mature DCs, as immature DCs have low to absent PD-1 expression. Furthermore, in the study done by Park et al., PD-1-deficient DCs led to improved overall T-cell immunity. This may be an important consideration for the use of anti-PD-1 to augment DC-based immunotherapies.

Intravital imaging and single-cell RNA sequencing done by Garris and colleagues (94) helps to provide insight on anti-PD-1 interactions with T cells and DCs. They found the robust immune response provided by anti-PD-1 therapy is largely due to T-cell and DC cross talk. This study by Garris et al. (94) underlines the importance of tumor-infiltrating dendritic cells (DCs) and shows that while tumor-infiltrating DCs do not directly bind to anti-PD-1 antibodies in vivo, they are critical in the licensing antitumoral T-cell activity. This is driven by downstream mechanisms where stimulated DCs and T cells drive production of IFN-γ and IL-12.

Anti-PD-1 directly binds and leads to the activation of T cells, which in response upregulates IFN-γ production sensed by tumor-infiltrating dendritic cells (DCs). In turn, DCs are stimulated and activated by IFN-γ, which induces DCs to increase IL-12 production through an indirect drug mechanism of action. IL-12 produced by DCs is critical in licensing T cells in exerting a robust immune reaction against the tumor mass (94). Activated DCs producing IL-12 are also enriched for NF-κB signaling proteins responsible in modulating effector immune response. IL-12 producing DCs also express increased costimulatory ligands responsible for provoking effector T-cell activity (94). IL-12 production by DCs has other downstream antitumoral implications like the promotion of T-bet in T cells and repression of eomesodermin (Eomes), a transcription factor responsible for immune exhaustive phenotype. T-bet is a transcription factor expressed in both innate and adaptive immune cells, required for immune cell survival, development, and effector function. Together T-bet and Eomes enhance IFN-γ production, affecting maturation of immune cells (98). Although anti-PD-1 antibodies are not known to directly bind to DCs as shown by Garris et al., DC and T-cell interactions are heavily influenced by the PD-1 checkpoint blockade. This is well corroborated by many including Krempski et al. who showed PD-1-expressing tumor-infiltrating DCs have immune-suppressive properties, leading to reduced T-cell activity and overall decrease in tumor-infiltrating T cells (94, 97, 99).

ANTI-PD-1 ANTIBODY EFFECT ON IMMUNE CELL CROSS TALK

Studies surveying tumor-infiltrating immune cells, note the involvement of three major subsets of immune cells: natural killer (NK) cells, macrophages, and dendritic cells (DCs) (Table 2) (53, 76, 77, 84, 104, 105). Activation of PD-1 in these cells have been found to inhibit immunity against tumors (75, 91, 93, 96, 102, 106). Anti-PD-1 treatment administered in mice knockout studies, and cell depletion of these immune subsets indicate macrophages and dendritic cells to be critical in anti-PD-1 drug efficacy (77, 84, 93, 94). NK cells were found to play an important role in amplifying immune antitumor response; however, they were not observed to be crucial for anti-PD-1 drug efficacy (84).

Table 2.

Effector function and immune stimulation by anti-PD-1 therapy

Cell Type	Effector Function
T cells	Enhanced proliferation of T-cell subsets (100) Increased expression of effector cytokines by infiltrating T cells (100) Elevated expression of IFN-γ and TNF-α (100) Increased tumor-infiltrating T cells (101) T cell-mediated antigen-specific immune response (93, 94, 100, 101)
Macrophages/monocytes	Increased tumor infiltration TAMs (tumor-associated macrophages) (53, 77, 84) Increase ratio of M1/M2 macrophages associated with improved prognosis and reduced tumor burden (53, 77, 84) Increase in phagocytosis of tumor cells (53, 77) Increase in IL-12 production and STAT1 activation (53, 102) Increased IL-6 production (53, 103)
Dendritic cells	Increased expression of CD40/CD40L, enhancing DC survival and resistance to apoptosis (93) Increased DC longevity (63) Recruit and prime T cells and stimulate local antitumoral response (93, 94) Increased secretion of IL-12 in response to anti-PD-1-activated T cells (94) Connect adaptive and innate immune cross talk and license immune response to tumor (57, 70)
Natural killer cells	Increased tumor infiltration (84, 104) Restored cytotoxic function after immune-dampening tumor microenvironment effect (104) Increased proliferation and differentiation (104) Increased production of granzyme B, perforin, and IFN-γ (104)

DC, dendritic cell; STAT1, signal transducer and activators of transcription 1.

Cross talk between innate cells (DCs, NKs, and macrophages) and adaptive immunity (T cells) corresponds to a robust immune response against tumor growth (76, 77, 84, 89, 93, 94, 103, 104, 106). Cross talk is predominantly facilitated by the secretion and production of cytokines. Active cytotoxic T cells secrete several cytokines, one of which is IFN-γ (94). Upon sensing IFN-γ, innate immune cells, dendritic cells, macrophages, and natural killer cells acquire a polarized cytotoxic immunological profile and secrete IL-12, which feedback to revitalize and engage T-cell cytotoxicity (Table 2). These events and interactions are critical in the engagement of antitumoral immunity and therapeutic efficacy of immune-modulating therapies like PEM.

Tracking data show that anti-PD1 antibody binds to T cells as early as 5 min after injection and has the greatest effect in the activation and cytotoxicity of T cells (89). However, long-standing sustained response and drug efficacy is facilitated by innate immune cells like dendritic cells and macrophages (53, 75, 93, 94, 106). Several studies have shown that although direct anti-PD-1 antibody interactions with innate cells is minimal, indirect consequences of this drug are critical in manifesting an overall therapeutic effect (84). Unfortunately, understanding this aspect of PEM therapy is limited. Further evaluations of immune activation and molecular pathways engaged by PEM therapy can help us to modulate immunity effectively, bypass tumor rejection, and prevent irAEs and conditions leading to serious chronic complications like CVD.

ANTI-PD-1 ANTIBODY EFFECT ON INTERCELLULAR COMMUNICATION AND CARDIAC DYSFUNCTION

PD-1 inhibitor-induced cardiotoxicity is preceded by increased immune infiltration of the heart by predominantly T cells and macrophages (107). Molecular analysis and bulk RNA sequencing shows PD-1 inhibition provokes M₁ macrophage polarization and T cell-mediated inflammatory signaling (107).

PD-1 inhibitor treatment of circulating monocytes and macrophages secrete microRNA-35a, which is transferred by exosomes and modulates cardiac injury (107). In cardiomyocytes, microRNA-35a induces senescence by telomere shortening and hindering cell cycle progression. Senescence leads to impaired cardiac tissue regeneration and reduced cardiomyocyte proliferation (107). Moreover, KLF4 was identified to be a target of microRNA-34a. KLF4 is an important cardioprotective transcriptional regulator, heavily involved in regulating immune balance in stress-induced inflammation (108). Overall, microRNA-34a secreted by exosomes produced by macrophages triggers senescence and hinders cardiac-regenerative capacity (108). This sensitizes tissues to inflammation while concurrently enhancing inflammation. This consequently leads to contractile dysfunction and heart failure.

Increased cytokine production by activated T cells (CD4⁺ and CD8⁺) has been observed to lead to cardiac dysfunction before immune cell infiltration of the myocardium and induction of cytokine-induced nitrosative stress (109). In T cells, IL-17 and CXCL9 have been reported to be differentially expressed and highly involved in mediating cardiac complications. IL-17 is associated with cardiac dysfunction after myocarditis detected in patients with heart failure (109). Anti-PD-1 treatment alters IL-17 (proinflammatory) and IL10 (anti-inflammatory) ratio inducing systemic inflammation, a primary culprit for development of irAEs (109). Recent work has shown that IL-17 blocking antibodies when used as an adjuvant can alleviate cardiotoxicity without disrupting anticancer properties provided by immune inhibition therapy (109). CXCL9 was also found to be significantly upregulated in the heart of post PD-1 inhibitor therapy. CXCL9 is known to increase chemotaxis, leukocyte extravasation, and induce CXCL9-CXCR3 signaling, and it is implicated in the pathology for heart failure (109). Clinical studies have reported higher levels of CXCL9 in patients with cancer suffering from irAEs receiving immunotherapies compared with those without irAEs (109).

Chronic inflammation triggers off-target side reactions (110). Posttreatment with anti-PD-1, results in widespread lymphocyte infiltration in tissues like the heart, leading to cardiomyocyte cell death and increased cardiac damage biomarker expression (111–115). The mechanism of anti-PD-1-associated cardiac dysfunction is not well understood. The occurrence of cytokine release syndrome (CRS) is presumed to be a primary culprit (3, 116). In CRS, immune cells (T cells, DCs, macrophages, and other innate immune cells) release cytokines (IL-6, IFN-γ, and TNF-α) and chemokines [reactive oxygen species (ROS) and nitric oxide (NO)] that have a tremendous effect on cardiomyocytes (3, 6, 51, 116–120). These factors result in abnormalities and dysfunction of the cardiovasculature, resulting in the development of arrhythmias, defects in cardiac contractility, and destruction of cardiac tissues. This is evident in various preclinical and clinical models of anti-PD-1-induced CVD (51, 111–113).

Animal models of PD-1 inhibition has been reported to induce cardiomyocyte apoptosis, autophagy, and necrosis (111–115). Furthermore, use of a cytotoxic T-lymphocyte antigen-4 (CTLA-4) antibody, another antibody-targeted immune inhibitory pathway, leads to increased expression of NLRP3 and MyD88, suggesting induction of inflammation and possible activation of pyroptosis cell death of cardiomyocytes exposed to immunotherapy (121, 122).

Anti-PD-1 therapy in tumor-bearing mice was investigated by Tay et al. (115). They found these anti-PD-1-treated mice exhibited significantly reduced heart function and dilated left ventricle. Further histological and gene expression studies of the heart tissue revealed significant infiltration of T lymphocytes in the myocardium and increased expression of IFN-γ, TNF-α, and apoptosis facilitator caspase-3 (115, 123).

While PD-1 immune checkpoint therapy is advantageous in eliciting immune response against tumors, it also induces severe immune reactions that severely impacts cardiomyocytes (Fig. 4). These reactions are primarily attributed to the production of cytokines and chemokines, which trigger cell death pathways (autophagy, apoptosis, pyroptosis, and necrosis) in heart, consequently leading fatal cardiac complications (122, 123). As such, patients undergoing immunotherapies must be monitored closely for the development of cardiac abnormalities so action can be taken to temper immune reactivity.

Figure 4.

Pembrolizumab cardiotoxicity–pembrolizumab blocks PD-1 and PD-L1/PD-L2 interactions, invigorating T-cell activity and inducing secretion of cytokines and chemokines to stimulate inflammation. While active immunity is beneficial in tumor clearance, these immune reactions also affect immune sensitive cardiomyocytes. Immune infiltration and inflammation induce cardiomyocyte cell death and damage to the myocardium. This affects contractility of the heart leading to fatal consequences, heart failure, and myocardial infarction. NO, nitric oxide; PD-1, programmed cell death protein 1; PD-L1/PD-L2, PD-1 ligand; ROS, reactive oxygen species; TCR, T-cell receptor.

ANTI-PD-1 ANTIBODY EFFECT ON OXIDATIVE STRESS AND INFLAMMATION

It is well established that immune cells play a major role in oxidative stress and inflammation in the tumor microenvironment (TME), which furthers disease progression (124). Much like immune surveillance of tumor growth, high reactive oxygen species (ROS) is highly effective in killing cancer cells, a common strategy used by chemotherapy (125). ROS has profound effects on cellular and organelle dysfunction and damage to DNA (126). Thus, we sought to understand the influencing role oxidative stress has on anti-PD-1 checkpoint blockade in the tumor microenvironment (TME).

Both a reactant and a product of inflammation, oxidative stress mediators—ROS and nitric oxide (NO)—play an important role in tumor development and immune regulation (3, 120, 127). It has been reported that high ROS and low antioxidant levels in subjects with cancer, corresponds with poor prognosis and increased refractory and recurrent tumors (3, 120, 127). Chronic exposure to high ROS levels are highly mutagenic and lead to genomic instability, concurrently facilitating the transition from tumor equilibrium to tumor escape by inducing exclusive survival of cell populations robust enough to adapt to high ROS stress (128–131). In addition, these cancer cells secrete potent cytokines (IL-6, IFN-γ, TGF-β, VEGF, IL-4, and IL-10) which over time, leads to immunosuppression of macrophages, dendritic cells, and T cells (51, 131).

Numerous in vivo and in vitro studies have reported oxidative stress and ER stress leading to increased tumor volume and accelerated cancer cell growth, with infiltration of increased dysfunctional and immunosuppressive immune cells like MDSCs (132). High ROS polarizes macrophages (the most abundant immune cell type in the TME) to M₂, which aids in tumor growth, metastasis, angiogenesis, and immunosuppression (110, 132). These macrophages support differentiation of regulatory DCs and MDSCs promoting immunosuppressive TME (132). Furthermore, MDSCs use immune checkpoint pathways like PD-1 to further deactivate immune cells like T cells and NK cells (133). Excess ROS and ER stress response severely undermines immune response to tumor growth. A study by Wang and colleagues (133) showed that the oxidative stress mechanisms in the TME leads to upregulation of T-cell apoptosis. High population of apoptotic T cells in the TME leads to the formation of larger tumor volumes and was further found to abolish therapeutic effect of PD-1/PD-L1 checkpoint blockade (133). This indicates the need for the use of antioxidants to bolster the potency of PEM.

PD-1 THERAPY IN CARDIAC DISEASE

Cardiovascular disease (CVD) and cancer account for the highest mortality and morbidity in the world (134). As population increases, this disease burden is also expected to follow. Breakthrough advancements in cancer management has enabled patient care to shift from treatment of cancer as a terminal condition to a chronic illness (134). This in combination with the prevalence in aged individuals with preexisting cardiovascular risk has contributed to increased occurrence of CVD and cancer. Both CVD and cancer originate from a state of uncontrolled inflammation (134). This is typically initiated by aberrancies in metabolism or inflammatory agents. Due to this overlap of inflammation-mediated progression of disease, cancer and cardiovascular disease are comorbidities intrinsically connected with shared risk factors. As a result, it has been observed that there is an increased risk of malignancy in individuals suffering from CVD as well as increased propensity for the development of CVD in patients with cancer (134). Inflammation from cardiac complications enhance tumorigenesis (135–137). In a study by Koelwyn et al. (136), myocardial infarction (MI) was observed to influence immune cell interactions in the TME. Post-MI, both tumor-bearing and nontumor-bearing mice acquire increased circulating immunosuppressive monocytes (135–137). These immunosuppressive monocytes were found to aid in tumor evasion and angiogenesis and supported tumor outgrowth stages: proliferation, migration, invasion, and metastasis. RNA sequencing of immunosuppressive monocytes showed differential expression and downregulation of gene pathways involved in immune restriction of tumor outgrowth. Immunosuppressive monocytes were reported to further influence tumor-infiltrating T cells, restricting their activity against tumor growth through PD-1/PD-L1 engagement. Overall, inflammation propagated by MI promoted breast cancer outgrowth. This emphasizes the importance of intercellular communications between immune cells in the heart and TME, as they have a wide range effect in CVD and tumor growth.

As previously discussed in the context of immunoediting, chronic inflammation imposes a selective pressure that shapes the survival of robust tumor cells by selecting highly metastatic and aggressive phenotypes (110, 117). The proinflammatory microenvironment gives rise to DNA damage and genetic instability that encourages acquired oncogenic characteristics like resistance to apoptosis and stimulation of angiogenesis (117). Mechanistically, inflammation drives production of ROS and NO inducing mutations and dysregulation of cell survival balance regularly conserved by tumor suppressor genes and protooncogenes. The inflammatory state induced by cancer cells can then further aggravate underlying CVD pathology (51, 58, 123, 134).

Alternatively, in CVD conditions, oncogenesis can be promoted. In myocardial infarction, necrotic and dying cells of the myocardium release DAMPs (damage-associated molecular patterns) that trigger and contribute to an intense inflammatory response. Furthermore, cardiac damage-induced stress like hypoxia leads to epigenetic response changes in DNA methylation and histone modifications corresponding to the development of malignancies. In hypoxia conditions induced by CVD, for example, activation of hypoxia-inducible factor 1 (HIF-1) pathway has been found to stimulate tumor growth and progression. Consequently, clinical studies have shown higher incidences of cancer in ischemic stroke survivors (134).

Clinical case studies have shown immune-based therapies to aggravate chronic inflammatory processes like atherosclerosis (58, 123, 138, 139). Atherosclerosis is a cardiovascular risk factor that gives rise to numerous cardiovascular events (51, 138). Disease progression involves chronic inflammation of arteries, where lymphoid and myeloid immune cells contribute to plaque and lesion formation that leads to MI or stroke (138, 139).

How immune therapies induce progression of atherosclerosis is unknown. However, much data exist to support that immune-based therapy aggravates the condition by increasing the number of infiltrating activated cytotoxic T cells in early atherosclerosis (58, 123, 138). In addition, it has been noted that there is an enhanced motility of T cells upon administration of immune-mediating antibody therapies. Increased expression of cell adhesion molecules such as intercellular adhesion molecule (ICAM) and vascular cell adhesion molecule (VCAM) in vascular endothelial cells was seen and may be the reason that increased T-cell infiltration in atherosclerotic plaques occurs (123, 138). Inflammatory mediators in the microenvironment of atherosclerotic plaques include inflammatory cytokines like IL-1, TNF-α, and IFN-γ. Many of which are produced by reactivated immune cells by immune therapies (51, 123, 134).

Clinical data to support the occurrence of immune-based therapy-induced atherosclerosis are limited, possibly because of the exclusion of patients with history of CVD in clinical trials (140). In addition, clinical complications of atherosclerosis are difficult to assess, as this condition develops very gradually over several years (138). Combating both cancer and CVD progression is critical not only for increased survival but also quality of life. Because of the interconnectivity of cardiovascular disease and cancer, a new branch of medical sciences known as cardio-oncology has been established and continues to play principal role in supporting patients who suffer from cancer and heart disease (134).

IMMUNOTHERAPY-INDUCED CARDIAC COMPLICATIONS

Although rare, cardiovascular-related side effects of immunotherapy are very severe and include the development of myocarditis, vasculitis, myocardial fibrosis, cardiomyopathy, and heart failure. The exact pathophysiology is not known but is thought to be caused by off-target immune response (140–142). Immune therapy complications in patients with cancer include acquired myasthenia gravis, myositis, and myocarditis with most reported early symptoms being fatigue and muscle weakness (143). Myocarditis is among the irAE associated with a 50% mortality rate (60). In addition, mouse models deficient in PD-1 were observed to acquire early cardiac dysfunction. Analysis revealed these mice produced autoantibodies to cardiac troponin I. In general, PD-1-deficient mice were predisposed to autoimmune disease and succumbed to fatal myocarditis by 10 wk of age. Further histological assessment showed extensive infiltration by T cells in the heart and, additionally, high amounts of autoantibodies against cardiac myosin (59). Further investigations revealed infiltrating CD8⁺ T cells have highly clonal TCRs with α-myosin epitopes, acting as an autoantigen responsible for the observed immunotherapy-induced cardiotoxicity (144).

Initial presentation of immunotherapy-related complications like myocarditis begins with general weakness and muscle pain (Table 3). Although reporting of these adverse reaction is rare, they are also underrecognized (160). It was reported that a patient who developed myogenic ptosis from PEM therapy consulted the ophthalmology department. Unfortunately, immunotherapy complications are not well known to ophthalmologists, and this patient’s condition quickly deteriorated and the patient passed away. This underlines the urgency for improved care and clinical monitoring for patients undergoing PEM therapy. Patients who develop myogenic ptosis and seek medical attention at eye clinics that are not equipped to manage immunotherapy complications, should be quickly transported to the ED (emergency department) for evaluation (153). These immunotherapy-induced neurological and myopathic symptoms has been observed to quickly progress to fatal cardiovascular and muscular complications (159). Fatal outcomes are attributed to myocarditis progressing to cardiac conduction block and myositis leading to respiratory muscle paralysis (162).

Table 3.

Clinical cases of PEM-induced irAEs

Clinical Case, Year Reported	Age/Sex (M/F)	PEM Oncological Indication	Reported PEM-Induced CVD Clinical Indications	PEM-Induced irAEs Clinical Presentation
Case 1, 2021 (145)	75	Malignant mesothelioma	Elevated CK, CK-MB, troponin-T ST segment elevation Severely reduced EF%	Myocarditis Hemodynamic compromise
Case 2, 2021 (60)	49/F	Thymoma	Elevated CK, troponin-T	Dysarthria Dysphagia Dyspnea Limb weakness Myocarditis
	70/M	Bladder cancer	Elevated CK, troponin-T	Ptosis Dyspnea Diplopia Muscle pain
	75/M	Chondroma of SACRUM	Elevated CK, troponin-T	Ptosis Diplopia Lower extremity edema
Case 3, 2021 (146)	83/M	Advanced nonsmall cell lung cancer	Elevated CK-MB, troponin-I, C-reactive protein ST-segment depression	Chest pain Dyspnea
Case 4, 2020 (147)	79/M	Metastatic melanoma	Elevated CK, troponin	Fatigue Lower extremity weakness Complete heart block Pulmonary embolism
	69/M	Metastatic urothelial carcinoma	Elevated troponin EF% of 40%–45%.	Diffuse body pain and weakness Concentric hypertrophy NSTEMI Stroke Hemiparesis
	89/M	Nonsmall cell lung cancer	Elevated CK, troponin EF% of 47%	Ventricular tachycardia Atrioventricular block Myositis Persistent weakness Dysphagia, blurred vision
Case 5, 2020 (148)	75/M	Lung adenocarcinoma	Elevated CK, CK-MB, troponin-I Extensive infiltration of CD8+ T cells and histiocyte infiltrates	Myalgia Palpitations Myocarditis Complete atrioventricular block Necrotizing myocarditis Necrotizing myopathy
Case 6, 2020 (149)	69/M	Urothelial carcinoma	Elevated CK, myoglobin, troponin-T Wide QRS complex ventricular rhythm Myocardial and muscle biopsy showed immune infiltration and inflammation in myofibers	Myalgia Severe fatigue and abnormal gait Reduced muscle reflexes and proximal muscle weakness Myositis proceeding to myocarditis
Case 7, 2020 (150)	77/M	Metastatic melanoma	Elevated CK, troponin, and creatine T-wave inversion	Acute weakness and fatigue Necrotic myopathy Myositis
Case 8, 2020 (151)	63/M	Bladder tumor	Elevated CK, CK-MB, troponin-I	Myositis Ptosis Diplopia Systemic myopathy
Case 9, 2020 (152)	70/F	Thymic carcinoma	Elevated CK, CK-MB, troponin-T T-wave inversions ST elevation Complete heart block Infiltrating lymphocytes and macrophages Early interstitial fibrosis and endocardial fibrosis in areas of immune infiltration	Dyspnea, orthopnea, and weakness Myocarditis Complete atrioventricular heart block Left lower lobe pulmonary embolism Anterior myocardial infarction
Case 10, 2019 (153)	83/M	Urothelial carcinoma	Elevated CK, CK-MB, troponin-I EKG showed T-wave abnormalities and nonspecific intraventricular conduction block	Fatal myocarditis Complete heart block Toxic myopathy/myositis
Case 11, 2019 (154)	70/M	Squamous cell carcinoma of the right lung	Elevated CK, CK-MB, troponin-T LVEF of 29% Lymphocytic infiltration in myocardial biopsy	Fulminant myocarditis Refractory ventricular fibrillation and circulatory failure
Case 12, 2019 (155)	58/F	Metastatic thymoma	Elevated troponin LVEF < 30% Diffuse hypokinesia of the heart	Anterior myocardial infarction Acute heart failure Decompensated heart failure Ventricular arrhythmia
	30/F	Metastatic B3 thymoma	Elevated CK, troponin-I	Acute chest pain Proximal muscle weakness Myositis/myocarditis
Case 13, 2019 (156)	83/M	Metastatic melanoma	Elevated CK, CK-MB, troponin-T, BNP T-wave abnormality, wide QRS, atrioventricular block	Fatigue and muscle pain Bilateral ptosis Diplopia Weakness of neck flexor and extensor Bilateral thigh myalgia Dysphagia and dyspnea Myositis and myocarditis
Case 14, 2018 (157)	67/F	Multiple myeloma	Elevated CK, CK-MB, troponin-T ST-segment elevation Complete AV block LVEF below 30% Infiltration of cytotoxic T cells and macrophages Collagen fibrosis in the myocardium	Myositis Myocarditis Depressed ventricular contractility without focal hypokinesis Ventricular tachycardia
Case 15, 2018 (158)	79/M	Metastatic gastric adenocarcinoma	Elevated CK, CK-MB, troponin-I Necrotic myofibers and infiltration of T cells and macrophages	Bilateral asymmetric ptosis Bilateral ptosis Myositis and myocarditis Dysphagia and bradycardia Atrioventricular block
Case 16, 2017 (159)	78/M	Metastatic melanoma	Elevated CK Reduced muscle action potential amplitudes Necrotic muscle fibers with immune infiltration	Acute onset of bulbar weakness Dysarthria Bilateral asymmetric ptosis Bifacial paresis Ophthalmoparesis Bilateral proximal limb muscle weakness Dysphagia Myalgia Muscle weakness in extremities Necrotizing myopathy Myositis of the diaphragm Myocarditis
Case 17, 2017 (160)	82/M	Melanoma	Elevated LDH, CK, C-reactive protein	Ptosis and miosis Proximal muscle weakness of the lower limbs Neck and shoulder pain unresponsive to painkillers and muscle relaxants Dysphagia, stridor, hypokinesis
Case 18, 2016 (161)	86/F	Metastatic melanoma	Elevated CK Muscle biopsy revealed necrosis with T-cell infiltration	Fatigue Left ptosis Ophthalmoplegia Proximal bilateral limb weakness Dyspnea Necrotizing myositis

BNP, brain natriuretic peptide; CK, creatine kinase; CK‐MB, creatine kinase myocardial band; CVD, cardiovascular disease; EF, ejection fraction; EKG, echocardiogram; irAEs, immune‐related adverse events; LDH, lactose dehydrogenase; LVEF, left ventricular ejection fraction; NSTEMI, non‐ST‐elevation myocardial infarction; PEM, pembrolizumab.

Off-target consequence of immunotherapy, like fulminant myocarditis and necrotizing myopathy, is associated with poor clinical outcome (141, 142, 159). Because myocarditis is a diagnosis of exclusion, early detection is critical (1). Patients who develop these complications have elevated serum biomarkers like cardiac Troponin T and I (cTnT, cTnI), creatinine kinase (CK), CK-MB (creatine kinase myocardial band), brain natriuretic peptide (BNP), and C-reactive protein (Table 3) (1, 163, 164). Creatine kinase has been observed to be the first biomarker to increase after administration of PEM (156).

Fulminant myocarditis initially presents with signs of acute heart failure. Symptoms include chest pain, dyspnea, and arrhythmias leading to syncope and sudden death. The pathophysiology of immunotherapy-induced fulminant myocarditis is not known but evaluations done postmortem show high levels of leukocyte infiltration by T cell and macrophages in the myocardium (140). Immunotherapy-induced necrotic myopathy progresses in a similar manner where histological evaluation reveals myofiber necrosis, atrophy, and infiltration of CD8⁺ T cells and macrophages (143, 150).

Because the use of immunotherapy is focused on reinvigorating immunity, patients with CVD-associated clinical indications and risk factors should be closely monitored and receive personalized treatments. Suspected myocarditis and myositis can be detected from EKG (electrocardiogram), biomarker measurement, chest radiology, cardiac imaging, and histological assessment of tissue biopsy (1). Treatments for immunotherapy-induced immune reaction include administration of corticosteroids, mycophenolate, cyclophosphamide, plasmapheresis, and intravenous immunoglobin (143). Plasma exchange can help to rapidly remove autoantibodies, immune complexes, cytokines, and the PEM drug (145).

Immunotherapy treatment is highly effective in reducing tumor burden with some cases having complete elimination of metastatic cancer (159). Many patients are now prescribed two different immune targets to aggressively target cancer progression (165). PEM is given to patients with cancer in conjunction with chemotherapies (platinum and fluoropyrimidine), other immune checkpoint inhibitors (anti-CLTA4 and anti-PD-L1), and is currently being investigated to be given in combination with tyrosine kinase inhibitors and chimeric antigen receptor T cell (CAR-T) therapies (165–170).

However, use of two or more therapies has been shown to be highly correlated with the development of immunotherapy-associated cardiovascular complications (134). As depicted in Table 3, clinical observations of immunotherapy-induced CVD complications from a single dose of PEM alone was observed to rapidly progress to fatal complications. To prevent this occurrence, careful monitoring, recognition, and early detection are key in managing patients treated with immunotherapies.

CONCLUSIONS

Cancer cells take advantage of immune checkpoints by upregulating PD-L1 expression to increase PD-1-mediated immune suppression. This phenomenon has been reported by many including Hamanishi et al. (171) who showed patients diagnosed with ovarian cancer and highly expressed PD-L1 had significantly poor prognosis. Intervention of this system by PEM, an immune checkpoint (PD1/PD-L1) blockade therapy, has shown to have tremendous effects in promoting tumor regression. While it may be advantageous to therapeutically target reactivation of antitumoral immunity, this can lead to severe complications of immune aggression (172). Immune-related adverse reactions from immune checkpoint therapy, like PEM, have a rapid onset that progresses to fatal and debilitating outcomes. Common clinical symptoms of immunotherapy-induced immune reactions include myalgia, proximal limb weakness, and myasthenia symptoms. These symptoms may occur within one dose of PEM and almost all cases of irAE report elevated creatine kinase levels (143). Upon PEM administration, CVD-related symptoms such as arrhythmias and depressed ejection fraction have been reported (Table 3) (143). These symptoms often begin with myocarditis, which can be definitively diagnosed by endomyocardial biopsy for evidence of inflammatory cell infiltrates in the myocardium. Because the use of PEM is focused on reinvigorating immunity, patients with CVD risk factors and those who exhibit muscle, cardiac, or neurological symptoms should be treated immediately and closely monitored. This review hopes to bring awareness to physicians taking care of patients undergoing immunotherapy treatments like PEM and emphasizes the need for a balanced approach where CVD risk and tumor regression are both taken into consideration and managed.

GRANTS

This study was supported in part by National Institutes of Health (NIH) Grant 1R01DK120866-01 (to D.K.S.), internal endowed chair Advent Health support (to D.K.S.), and NIH Grant R01-CA-221813 (to D.K.S.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

D.K.S. conceived and designed research; A.B. and C.A.N. analyzed data; A.B. interpreted results of experiments; A.B. prepared figures; A.B. drafted manuscript; C.A.N. and D.K.S. edited and revised manuscript; D.K.S. approved final version of manuscript.