Immune Checkpoint Therapies and Atherosclerosis: Mechanisms and Clinical Implications: JACC State-of-the-Art Review

Abstract

Immune checkpoint inhibitor therapy has revolutionized the treatment of advanced malignancies in recent years. Numerous reports have detailed the myriad of possible adverse inflammatory effects of immune checkpoint therapies, including within the cardiovascular system. However, these reports have been largely limited to myocarditis. The critical role of inflammation and adaptive immunity in atherosclerosis has been well characterized in preclinical studies, and several emerging clinical studies indicate a potential role of immune checkpoint targeting therapies in the development and exacerbation of atherosclerosis. In this review, we provide an overview of the role of T-cell immunity in atherogenesis and describe the molecular effects and clinical associations of both approved and investigational immune checkpoint therapy on atherosclerosis. We also highlight the role of cholesterol metabolism in oncogenesis and discuss the implications of these associations on future treatment and monitoring of atherosclerotic cardiovascular disease in the oncologic population receiving immune checkpoint therapy.

Condensed Abstract

In the last decade, immune checkpoint inhibitors have been revolutionary in the field of oncology and has become the standard of care in numerous advanced malignancies. As T cell immunity plays an integral role in atherogenesis through promotion of inflammation, preclinical studies and recently emerging clinical studies indicate a role of immune checkpoint inhibitors in the development of atherosclerosis. This review provides a comprehensive analysis of the molecular effects and clinical associations between immune checkpoint altering therapy and atherosclerotic cardiovascular disease (ASCVD) risk, and identifies important considerations to the surveillance and treatment of atherosclerosis in patients receiving immune checkpoint therapy.

Introduction: Emergence of Immune Checkpoint Inhibitors

Modern day immunotherapy originates from the cancer immunosurveillance hypothesis, which states that immune cells are responsible for the surveillance and elimination of nascent transformed cells in host tissues(1). Enhanced appreciation for the ability for tumor cells to evade immune surveillance (2) galvanized efforts to develop therapies restoring the host immune system’s antineoplastic defenses. The culmination of these efforts led to the development of immune checkpoint inhibitors (ICIs) that restore the host T cell immune response against cancer cells. The development of ICIs in the last decade has rapidly revolutionized cancer therapy, with applications of ICIs targeting CTLA-4 and PD-1/PDL-1 pathways in cancers including melanoma, non-small cell lung cancers (NSCLC), and urothelial carcinomas(3-5).

When ICIs were being approved, it was anticipated that diffuse leveraging of the immune system would lead to off-tumor adverse events, which occur in up to 70% of patients and may affect any organ system, but are typically easily managed. Initial reports detailing the potential cardiovascular effects of ICIs focused principally on the uncommon, but highly morbid, occurrence of myocarditis, mediated by direct T cell infiltration of the myocardium expressing PD-L1(6). More recently, the potential for ICIs to affect the cardiovascular system beyond myocarditis has been reported. Specifically, preclinical evidence supports that treatment with ICIs can theoretically promote the development and acceleration of atherosclerosis, and recent emerging clinical evidence suggests an increase in atherosclerotic-related cardiovascular events in patients receiving ICI therapy and a possible connection between ICIs and increased atherosclerotic cardiovascular disease (ASCVD) risk (7).

The role of inflammation in atherosclerosis has been well established, with the CANTOS, LoDoCo, and COLCOT trials demonstrating that targeting immune responses can improve clinical cardiovascular outcomes(8-10). These studies showed that targeting innate immunity, such as inhibition of IL-1β via canakinumab and neutrophil function via colchicine, may mitigate the progression of ASCVD. As adaptive immunity via T cell activation has also been shown to play a crucial role in the development and progression of atherosclerosis, ICIs may have important implications on atherosclerotic disease (11). This review aims to describe the role of T-cell mediated immunity in atherogenesis, the molecular implications of various pathways of immune checkpoint alteration in atherosclerosis, the current clinical associations between treatment with ICIs and atherosclerosis, and potential treatment avenues.

Overview of T Cell Mediated Immunity and Key Immune Checkpoint Pathways

1. Overview of T Cell Mediated Immunity

Tumor associated antigens are recognized and phagocytosed by antigen presenting cells (APCs) such as dendritic cells and macrophages. These antigens are presented via major histocompatibility complex (MHC) molecules on the surface of APCs, which interact with T cell receptors (TCRs) and serve as the primary stimulatory signal leading to the intracellular cascade that activates naïve CD4+ and CD8+ T cells. However, full T cell activation is dependent on the presence of co-stimulatory signals involving the binding of CD28 on T cell surfaces to B7-1 (CD80) or B7-2 (CD86) on APCs (Figure 1A).

Figure 1. T Cell Activation and Effect of Immune Checkpoints.

A. T Cell Activation. Antigen presenting cells (APCs) present antigens via major histocompatibility complex (MHC) molecules that bind to T cell receptors (TCR) on naïve CD4+ and CD8+ T cells. T cell activation requires the binding of costimulatory molecules B7.1(CD80) or B7.2(CD86) on APCs to CD28. Naïve CD8+ T cells are activated into cytotoxic T cells that secrete perforins and granzymes, leading to apoptotic cascades in target cells. The presence of particular cytokines and growth factors in the microenvironment determine naïve CD4+ T cell differentiation.

B. Immune Alterations in Cancer and Effect of Immune Checkpoint Inhibitors. Prolonged inflammation in cancer leads to T cell exhaustion that promotes recruitment of T_reg cells and increased expression of cytotoxic T-lymphocyte associated protein 4 (CTLA-4) and programmed cell death protein 1 receptor (PD-1) on T cells. Programmed cell death protein 1 receptor ligand (PD-L1) on tumor cells binds to PD-1 to decrease the production of inflammatory cytokines. Immune checkpoint inhibitors enhance tumor cell killing by targeting PD-1, PD-L1 and CTLA-4 to restore T cell activation and inflammatory cytokine production. Created with BioRender.

Subsequently, the cytokine composition of the surrounding environment determine the fate of T cell differentiation into various subclasses, including CD8+ cells into cytotoxic T cells, and CD4+ T cells into helper T cells (T_h1, T_h2 , and T_h17) and regulatory T cells (T_reg) (Figure 1A). Cytotoxic T cells secrete cytotoxins and promote apoptosis of its target cells. Of the CD4+ T cell derivatives, T_h1 promotes cell-based immunity by activating macrophages and cytotoxic T cells, while T_h2 cells promote antibody-mediated immunity and the recruitment of eosinophils. T_h17 cells assist with extracellular pathogen clearance at mucosal surfaces and promote antibody production (12). T_reg cells promote peripheral tolerance by secreting inhibitory cytokines, promoting cytolysis, and suppressing the activation, proliferation and cytokine production of CD4+ and CD8+ T cells (13).

2. T Cell Anergy and Co-Inhibitory Signaling

Several mechanisms exist to prevent immune overactivation and promote self-tolerance. Activation of T cells requires binding of a second co-stimulatory signal with stimulatory checkpoint molecules such as CD28 to B7-1/B7-2, without which T cells would remain anergic. Another class of immune checkpoint molecules induce co-inhibitory signals, including cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and the programmed cell death protein −1 (PD-1) receptor.

CTLA-4 is an analog similar to CD28 that is constitutively expressed on T_reg cell surfaces and upregulated on other T cell classes after initial activation. It exerts inhibitory effects in the early phases of T cell activation within lymphoid tissues by directly competing with CD28 to bind to B7-1 and B7-2 ligands on APCs with higher affinity. Its binding launches a signaling cascade preventing TCR signal transduction (14). Later studies showed a signaling-independent mechanism of inhibition via activation of trans-endocytosis that removes B7 ligands from the surfaces of APCs (15). CTLA-4 expression has been demonstrated in T cells of atherosclerotic plaque (16), and CTLA-4 inhibition has been linked to multiorgan lymphocyte infiltration, including the heart, leading to ICI fulminant myocarditis (17).

The PD-1 receptor is present on activated T cells and exerts an inhibitory effect via binding its two ligands PD-L1 (CD274) and PD-L2 (CD273) in peripheral tissues. While PD-L2 is present mostly on macrophages and dendritic cells, PD-L1 is expressed in hematopoietic cells and tissue cells in various organs, including in cardiomyocytes, vascular endothelial cells, and leukocytes within atherosclerotic plaque (18-20). The binding of PD-1 to PD-L1 or PD-L2 leads activation of the Akt pathway that decreases the production of inflammatory cytokines and cell survival proteins (21). Disruption in the PD-1/PD-L1 pathway in PD-1 knockout mice led to rapid onset autoimmune mediated dilated cardiomyopathy with diffuse IgG deposition, and is a potential mechanism of cardiotoxicity in ICI induced myocarditis (19).

3. Immune Checkpoint Alterations in Cancer and the Advent of ICIs

In the tumor microenvironment, tumor cells promote recruitment and production of CTLA-4 expressing T_reg cells via TGF-β secretion. (22). In addition, prolonged activation of T cells in the cancer immune response leads to exhaustion, where T cells upregulate inhibitory molecules including PD-1 and CTLA-4 to limit their own proliferative potential (23) (Figure 1B). The surrounding tissue hypoxia promotes an increase in PD-L1 expression in various tumors (18).

Targeted ICIs utilize the altered oncologic expression of immune checkpoints to augment the host immune response, beginning with the approval of the CTLA-4 checkpoint inhibitor ipilimumab (Yervoy) for use in metastatic melanoma in 2011 (4). Shortly after came approval for PD-1 inhibitors pembrolizumab (Keytruda) and nivolumab (Opdivo) (3, 5) (Figure 1B). Since their initial FDA approval, the indications for ICIs have expanded rapidly to include NSCLCs, non-Hodgkin’s lymphoma, urothelial carcinoma, colorectal cancers, and more. Several investigational therapies targeting other immune checkpoints are in various stages of development (24). The pathways utilized in anti-CTLA-4 and PD-1/PD-L1 inhibitor therapies as well as those of investigational immune checkpoint agents have been implicated in atherogenesis (Central Illustration, Table 1).

Central Illustration: Summary of Immune Checkpoint Alterations and their Effects on Atherogenesis.

CTLA-4 and PD-1/PDL-1 blockade are well-studied examples of co-inhibitory molecule blockade that worsen atherosclerosis. CD40/CD40L agonism and GITR agonism are examples of co-stimulatory molecule agonism that worsen atherosclerosis. CD-47 blockade restores efferocytosis and may be atheroprotective. APC indicates antigen presenting cell; MHC, major histocompatibility complex, TCR, T cell receptor, CTLA4, cytotoxic T-lymphocyte associated protein 4; PD-1, Programmed cell death protein 1 receptor; PD-L1, Programmed cell death protein 1 receptor ligand; SIRP-α, signal-regulatory protein alpha; CD, cluster of differentiation; GITR, glucocorticoid-induced tumor necrosis factor family-related protein; GITRL, glucocorticoid-induced tumor necrosis factor family-related protein ligand; Treg, regulatory T cell; VCAM, vascular cell adhesion molecule; and MMP, matrix metalloproteinase. Created with BioRender.

Table 1:

Summary of Immune Checkpoint Pathway Alterations and Effects on Cancer Therapy and ASCVD

Immune Checkpoint Class	Immune Checkpoint Target	Effect on Cancer Therapy†	Effect on ASCVD†	Example Therapies
Co-Inhibitory Signal Blockade	CTLA-4	Restores T cell activation and enhances cancer cell destruction (3, 24)	-Increases atherosclerotic lesion size (63-65) -Progression to advanced phenotype (65) -Decreases collagen content (65)	Ipilimumab (4) Tremelimumab (151)* Zalifrelimab (AGEN1884)* (152)
	PD-1 and PD-L1		-Increases atherosclerotic lesion size (67-69) -Enhances TNF-α secretion (67, 68) -Enhances lesion helper T cell, cytotoxic T cell and macrophage activation (67-69)	Anti-PD-1: nivolumab (5), pembrolizumab (3), cemiplimab (153), spartalizumab* (154), camrelixumab* (154), tislelizumab* (154) Anti- PD-L1: atezolizumab (155), avelumab (156), durvalumab (157), cosebelimab*(154),sugemalimab*(154), CX-072*(154)
	TIM-3*		-Increases atherosclerotic lesion size (105) -Increases lesion macrophage content -Decreases lesion Treg content -Enhances TNF-α and IFN-γ secretion in combination with anti-PD-1 in human cells (106)	Cobolimab* (24) BMS-986258*(158), MBG453* (158), TSR-022*(158), Sym023*(158), BGB-A425* (158)
	LAG-3*		-Effect unknown, a marker positively correlated with ASCVD in a human observational study (109)	Relatlimab*, Eftilagimod alpha*, Tebotelimab*, Favezelimab*, LAG525*, REGN 3767* (159)
Co-Stimulatory Signal Agonism	GITR*	-Enhances macrophage activity (90) -Promotes CD4+ and CD8+ T cell activation (89, 90) -Decreases immunosuppressive ability of Treg (89)	-Promotes macrophage secretion of atherogenic cytokines (94, 95) -Increases MMP production (95) -Enhances plaque leukocyte recruitment via ICAM-1 (95) -GITR inhibition decreased plaque size and increased plaque stability (94)	AMG228*, MEDI-1873*, BMS-986156* (160)
	CD40 and CD40L*	-Enhances antigen presentation by APCs (74) -Promotes B cell proliferation and antibody production (74) -Promotes tumor cell apoptosis (75)	-Increases secretion of IL-1β, IL-6, TNF-α, IFN-γ (73) -Endothelial cell and smooth muscle cell activation in human cells (78) -Increases leukocyte recruitment by expression of VCAM-1 and P-selectin (77) -Increases MMP production (77) -Promotes foam cell production (77) Inhibition of CD40L: -increased plaque stability (82-85) - inconsistently decreased plaque size (82, 85) vs no change (83, 84) -reduced Th1 polarization and IFN-γ secretion (87) -reduced atherothrombosis (87)	Selicrelumab* (161), dazetuzumab* (162), CP-870893*(162) , JNJ-107*(162), APX005M (162)*
	CD27 and CD70*	-Enhanced T cell expansion and survival (96) -Enhanced memory T cell production (96) -Polarization toward IFN-γ producing effector T cells (96)	-decreased lesion size (98, 99) -enhanced macrophage efferocytosis (98) -enhanced oxLDL clearing (98) -promotion of Treg survival (98, 99)	Anti-CD27: Varlilumab*, AMG-172* Anti-CD70: Cusatuzumab*, BMS-936561* (163)
	ICOS and ICOSL*	-Promotion of cytotoxic T cells (100) -Enhancing effector T cell function (100) -Promotion of Treg activity (100)	Inhibition of ICOS: - increased atherosclerotic plaque size (101, 102) -reduced atherosclerotic Treg population (102) -increased IFN-γ secretion (101) -decreased IL-10 secretion (101)	Anti- ICOS Agonists: Feladilimab*,vopratelimab (100) Anti- ICOS Antagonists: MEDI-570*, KY1044 *(100)
“Don’t eat me” Blockade	CD47*	-Blockade of interaction with SIRP-α (111) -Enhances phagocytosis of apoptotic cells and debris (111) -Reduces inflammation (111)	-Reduces atherosclerotic lesion size and inflammation (112, 113) -Decreases macrophage response to IL-1β and IFN-γ (113) -Improves clearance of VSMCs (114)	Magrolimab* (115), SRF231*(164), AO-176*(164), CC-90002* (164)

^*denotes investigational therapies.

^†denotes findings of preclinical/animal studies unless otherwise specified. AGEN indicates Agenus; AMG, Amgen; AO, Arch Oncology; APX, Apexigen; APCs, antigen presenting cells; ASCVD, atherosclerotic cardiovascular disease; BGB-A, BeiGene; BMS, Bristol Myers Squibb; CC, Celgene; CD, cluster of differentiation; CTLA-4, cytotoxic T-lymphocyte associated protein; CX, CytomX; GITR, glucocorticoid-induced tumor necrosis factor receptor-related protein; ICAM-1, intracellular adhesion molecule 1; IFN-γ, interferon gamma; IL, interleukin; JNJ, Johnson & Johnson; KY, Kymab; LAG-3, lymphocyte-activation gene 3; MMP, matrix metalloproteinase; PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; REGN, Regeneron Pharmaceuticals; SIRP-α, signal regulatory protein alpha; SRF, Surface Oncology; Sym, Symphogen; TIM-3, T cell immunoglobulin and mucin-domain-containing-3; TNF-α, tumor necrosis factor alpha; T_reg, regulatory T cell; VCAM-1, vascular cell adhesion molecule 1; and VSMCs, vascular smooth muscle cells.

Role of Immune Cells in Atherosclerosis and Effects of Current Immune Checkpoint Inhibitors

1. Overview of Atherogenesis

The pathophysiology of atherosclerosis begins with damaged endothelial cell expression of cellular adhesion molecules that attract monocytes to enter the subendothelium, where they transform into macrophages that are polarized into subtypes depending on their microenvironment. The classically activated pro-inflammatory phenotype is promoted by exposure to free fatty acids, oxidized lipids, and factors such as IFN-γ whereas the alternatively activated anti-inflammatory phenotype is promoted by factors such as IL-4, IL-13, and IL-10. Pro-inflammatory macrophages predominate in early atherosclerotic plaque and are responsible for the accumulation of intracellular lipids, formation of foam cells, and secretion proinflammatory cytokines such as IL-1β and IL-6 (25) (Figure 2). Anti-inflammatory macrophages promote collagen formation and effective lipid clearance, and are associated with atherosclerosis regression (26). Necrotic core formation is the hallmark of chronic inflammation in atherosclerosis, and is rich in lipids and cellular debris as a result of foam cell and vascular smooth muscle cell (VSMC) apoptosis and necrosis (27).

Figure 2. Atherogenesis and Effects of T Cell Activation.

Damaged endothelial cells express platelet endothelial cell adhesion molecules (PECAM), intracellular adhesion molecules (ICAM), vascular cell adhesion protein (VCAM) and selectins that recruit monocytes to the subendothelium. Monocytes are activated into macrophages that consume oxidized low density lipoproteins (ox-LDL) and produce inflammatory cytokines, leading to foam cell and necrotic core formation. T cells are recruited to the subendothelium and activated by APCs. Activation of T helper 1 (T_h1) cells leads to tumor necrosis factor alpha (TNF-α) secretion that promotes monocyte recruitment and endothelial damage, and interferon gamma (IFN-γ) secretion that promotes macrophage activation. T cell activation decreases the function of regulatory T cells (T_reg) that normally decreases T cell and monocyte recruitment via transforming growth factor-beta (TGF-β) secretion and reduces T_h1 differentiation via interleukin-10 (IL-10) secretion. T cell activation is enhanced by immune checkpoint altering antibodies. Created with BioRender.

T cell activation is also integral to atherogenesis. During early atherosclerosis, APCs present atheroma-related antigens to naïve T cells in lymphoid tissues, leading to T cell migration towards the plaque (28). The initial recruitment of T cells is nonspecific, and they undergo further selective activation and clonal expansion into atherogenic T cell subtypes (28, 29). Atheroma-related T cell antigens have been difficult to identify, but include oxidized LDL (oxLDL) particles, heat shock proteins, and Apolipoprotein B (30). OxLDL are abundant in atheromatous plaque, and serves as an antigen triggering T cell autoimmune response that leads to inflammation and macrophage activation via IFN-γ secretion (31). CD8+ cytotoxic T cells, T_h1 cells, T_h2 cells, T_h17 cells, and T_reg cells have all been identified in atherosclerotic plaques (30).

2. Role of Differentiated T cells in Atherogenesis

T_h1 cells are the predominant CD4+ T cell present in plaques (32) and the most directly associated with atherogenesis due to their production of inflammatory cytokines, including IFN-γ and TNF-α (33, 34) (Figure 2). The production of IFN-γ enhances recruitment of macrophages and T cells and promotes macrophage polarization, cytokine secretion and foam cell formation (35). IFN-γ is also thought to inhibit vascular smooth muscle cell proliferation, thereby contributing to decreased plaque stability (36). The atherogenic role of IFN-γ is most clearly demonstrated in a study where mice with dysfunctional IFN-γ receptors developed smaller and more phenotypically stable atheromas compared to controls (37). TNF-α promotes atherosclerosis through leukocyte recruitment, inflammatory cytokine production, and promotion of endothelial cell damage and oxidative stress (38). TNF-α deficient mice have been shown to exhibit smaller plaque lesions (33), and the presence of TNF-α is associated with increased lesion necrosis and more advanced plaque progression in mice (39). The inhibition of T_h1 differentiation in mice is also atheroprotective and reduces the amount of IFN-γ detected in plaques (40).

T_reg cells’ atheroprotective role has been well studied, and their effects are primarily mediated through secretion of TGF-β and IL-10 (Figure 2). TGF-β inhibits recruitment and activation of T-cells and macrophages, and increases plaque stability by promoting VSMC proliferation (41). Mice with defective TGF-β receptors have larger atherosclerotic lesions with increased T cell and macrophage composition, increased IFN-γ expression, and more vulnerable plaque phenotype (42). IL-10 reduces T_h1 differentiation and prevents the recruitment and cytokine secretion of T cells and macrophages (43). A study with IL-10 deficient mice demonstrated increased susceptibility to atherosclerosis, and higher T cell infiltration and IFN-γ expression compared to controls (44). In addition, T_reg secretion of IL-10 promotes the transformation of macrophages from the pro-inflammatory to the anti-inflammatory phenotype, potentiating its atheroprotective effect (45). Ait-Oufella et al. demonstrated that deficiency in T_reg was associated with increased plaque size and more advanced plaque phenotype in mice, and that subsequent co-transference of T_reg reduced inflammatory cell infiltration and plaque size (46). In addition, the number of T_reg cells was inversely correlated with plaque vulnerability in human carotid arteries (47).

The roles of the remaining T cell subtypes in atherosclerosis are less well defined. T_h2 cells oppose the production of IFN-γ, which suggested a potentially atheroprotective role. This was further supported as deficiency in IL-5, one of the primary T_h2 secreted cytokines, was shown to accelerate atherosclerosis development in mice (48, 49). However, the deletion of another T_h2 cytokine, IL-4, has been shown to decrease plaque lesion size in mice (50). Thus, the role of T_h2 cells in atherogenesis remains unclear.

T_h17 cells have received considerable attention in atherosclerosis in recent years, though its definitive role is still controversial. A study analyzing human coronary atherosclerotic plaques demonstrated that IL-17, the principal cytokine released by T_h17, worked synergistically with IFN-γ to promote inflammation by increasing secretion of IL-6 (51). However, cytokine secretion of T_h17 cells relies on environmental context, and subsets secrete IL-17 in conjunction with the anti-inflammatory IL-10, making the role of T_h17 cells in atherosclerosis difficult to define (52). Several studies have indicated an atherogenic role to IL-17,(53) while others suggest an atheroprotective role and promotion of plaque stability via Type I collagen production (54, 55). Despite these inconclusive findings, the ratio between T_h17 and T_reg cells have been implicated in atherosclerosis progression, with increased T_h17 and decreased T_reg levels observed in patients with coronary atherosclerosis. Recent data suggests that inhibition of CD69, a key molecule expressed on early T lymphocytes regulating T cell differentiation, leads to elevated T_h17/T_reg ratios and exacerbation of atherosclerosis in mice (56). Indeed, the delicate balance between T_h17 and T_reg cells has important implications on autoimmune conditions and off-target adverse events related to ICI therapy, including myocarditis (57, 58).

Despite CD8+ T cells comprising about 50% of lymphocytes in atherosclerotic lesions (59), fewer studies have addressed the role of CD8+ cells. Cytotoxic CD8+ T cells have been shown to promote atherosclerosis in mouse models through IFN-γ secretion (60) and necrotic core formation by perforin and granzyme B-mediated apoptosis of macrophages (61). However, another mouse model suggest an atheroprotective role by promoting cytolysis of APCs (62).

Immune Checkpoint Therapies and Atherosclerosis

1. B7-1/B7-2 and CD28/CTLA-4 Pathway

The immunoregulatory effects of CTLA-4 and B7-1/B7-2 binding suggests an atheroprotective role for this pathway. The use of abatacept, a synthetic analog of CTLA-4, prevented CD4+ cell activation and reduced atherosclerosis development in murine femoral arteries by 78%, whereas the administration of CTLA-4 blocking antibodies increased atherosclerotic lesion sizes (63). Similarly, mice with elevated homocysteine levels had larger atherosclerotic plaque sizes associated with decreased membrane expression of CTLA-4, and pretreatment of these mice with abatacept ameliorated plaque development with reduced IFN-γ and IL-2 production and decreased macrophage content (64). Matsumoto et al. demonstrated that transgenic mice with constitutive CTLA-4 overexpression exhibited a significant reduction in atherosclerotic lesion size at the aortic root (16). In this study, CTLA-4 overexpression appeared to reduce atherosclerosis by decreasing plaque inflammation, as evidenced by a 38% decrease in macrophage accumulation and 42% decrease in CD4+ T cell infiltration, and downregulation of T cell proliferative capacity and proinflammatory cytokine production (16).

Modeling the role of ICIs in atherosclerosis, Poels et al. evaluated the role of antibody mediated CTLA-4 inhibition on atherogenesis. They demonstrated a two-fold increase in the size of atherosclerotic lesions in mice treated with CTLA-inhibiting antibodies, which was primarily mediated by a transition to an activated T cell profile without significant alterations in the macrophage inflammatory profile. CTLA-4 inhibition was also associated with plaque progression to more advanced phenotypes, with decreased collagen content and increased intimal thickening and necrotic core areas (65).

2. PD-1 and PD-L1/PD-L2 Pathway

PD-1 has been shown to suppress T_h1 cytokine production and promote the development of T_reg cells, suggesting a potentially atheroprotective role for this pathway (66). In hyperlipidemic mice, both PD-L1/PD-L2 deficiency and PD-1 receptor inhibition have been associated with increased atherosclerotic lesion size, increased plaque T cell activation , and enhanced TNF-α secretion (67, 68) . In contrast to CTLA-4 inhibition, PD-1 inhibition also exhibited increased lesion macrophage content, and enhanced the cytotoxicity of lesion CD8+ T cells (68). PD-1 deficiency has been shown to activate both CD4+ T cells and regulatory T cells, but the net effect was still exacerbated atherosclerotic lesion growth and increased plaque T cell infiltration (69).

In humans, several studies have observed decreased expression of PD-1 or its ligands in patients with coronary artery disease (CAD) and acute coronary syndrome (ACS), suggesting its protective role in atherogenesis and progression to advanced plaque phenotype (70, 71). At baseline, human atherosclerotic plaque T cells express high levels of PD-1 as a marker of T cell exhaustion in chronic inflammation (72). Thus, the delicate co-existence of PD-1 expressing T cells with activated T-cells within human plaques raises concern that PD-1 inhibition would lead to exacerbation of atherosclerosis.

Investigational Immune Checkpoint Targets and Atherosclerosis

1. CD40-CD40L Pathway

CD40 is a T cell costimulatory molecule and a member of the TNF receptor family that is present on APCs as well as non-immune cells such as endothelial cells, VSMCs, and platelets (73). Its classic ligand, CD40L, is expressed on activated CD4+ T cells. The binding of CD40 to CD40L promotes B cell proliferation and improves dendritic cell antigen presentation and T cell activation (74). CD40/CD40L ligation leads to recruitment of TNF receptor associated factors (TRAFs), that lead to downstream increases in atherogenic cytokines including IL-1β, IL-6, TNF-α and IFN-γ, which enhances the activities of macrophages (73).

Activation of the CD40/CD40L pathway has been shown to enhance antitumor immune response and promote tumor cell apoptosis and has been explored in cancer immunotherapy(75). Several ongoing Phase I and II clinical trials are evaluating the effect of anti-CD40 agonist antibodies, such as elicrelumab and dacetuzumab in various hematologic and solid tumor malignancies (76).

Several studies have explored role of the CD40/CD40L pathway in atherosclerosis. CD40L ligation to CD40 promotes thrombosis due to its activating effects on endothelial cells and platelets and its promotion of tissue factor production(77, 78). Endothelial cells exposed to CD40L exhibit increased expression of adhesion molecules such as VCAM-1 and P-selectin that enhances leukocyte recruitment, a key process in atherosclerosis initiation (79). CD40L has also been shown to promote foam cell formation in atherosclerotic plaques (80). Bruemmer et al. observed that in human iliac arteries, the expression of CD40 increased with atherosclerotic stage, and the largest increases occurred in the early stages of atherosclerosis (81). In later stages of atherosclerosis, the CD40/CD40L pathway stimulation increased secretion of matrix-degrading metalloproteinases that increase plaque rupture vulnerability (78).

Several preclinical studies have explored the inhibition of CD40/CD40L in reducing atherosclerosis. Blockade of CD40L either via antibodies or knock out models resulted in decreased leukocyte composition in the atherosclerotic plaque and phenotypic evidence of increased plaque stability (82-85). However, CD40L blockade was inconsistent in achieving plaque size reduction, where studies by Bavendiek et al. and Mach et al. observed reduced plaque size (82, 85), while two studies from Lutgens et al. did not (83, 84). In addition, CD40L inhibitory antibodies has been shown to increase the risk of thromboembolic events via platelet activation, limiting its systemic use in atherosclerosis treatment (86). Recent data from Lacy et al. demonstrated differing roles for CD40L depletion on atherosclerosis depending on cell source, with T cell CD40L inhibition leading to reduced T_h1 polarization and IFN-γ secretion, whereas platelet-CD40L inhibition led to reduced atherothrombosis (87). These data hold promise for the development of cell-specific CD40L inhibition in the treatment of atherosclerosis.

With further research, CD40L/CD40 blockade may be pursued as a treatment for atherosclerosis. Conversely, with CD40/CD40L agonists being studied in clinical trials for cancer immunotherapy, their possible implications on cardiovascular disease must be carefully considered.

2. GITR-GITRL Pathway

Similar to CD40, glucocorticoid-induced TNF receptor-related protein (GITR) is a T cell costimulatory molecule present on T and NK cells and binds to its ligand, GITRL, expressed on APCs and endothelial cells (88). GITR-GITRL signaling promotes macrophage activity, increases CD8+ T cell cytotoxicity, and enhances effector T cell activation by increasing IFN-γ and IL-2 secretion (89, 90). Simultaneously, T_reg cells experience a decrease in immunosuppressive effect downstream of GITR-GITRL interaction (89). The efficacy of tumor inhibition by GITR agonists was directly correlated with intratumor CD4+ and CD8+ levels, suggesting its therapeutic potential in NSCLC, renal cell carcinoma, and melanoma (91). Several phase I clinical trials have shown limited therapeutic response as monotherapy, but possible synergistic activity in combination with anti-PD-1 agents (92).

Macrophage and effector T cell activation and T_reg suppression by the GITR/GITRL pathway suggests a proatherogenic role. GITR+ macrophages, T cells, and endothelial cells have been identified within atherosclerotic plaques (93, 94). Moreover, Shami et al. demonstrated increased GITR expression in carotid artery plaques in patients with symptomatic cerebrovascular disease, which correlated with increased plaque vulnerability (94). Kim et al. showed that GITR agonist antibodies promoted macrophage production of pro-atherogenic cytokines, plaque-destabilizing metalloproteinases, and intracellular cell adhesion molecule 1 (ICAM-1) (95). Conversely, GITR deficiency has been associated with decreased plaque size and increased plaque stability in mice (94). Thus, GITR agonism may be beneficial in combination with other ICIs such as PD-1 in cancer immunotherapy, but may worsen atherosclerosis, though studies of the direct effect of GITR agonistic monoclonal antibody on atherosclerosis are needed.

3. Other Costimulatory Pathways: CD27/CD70 and ICOS

Agonism of several other co-stimulatory pathways are being explored as immunotherapy. CD27 is a transmembrane glycoprotein that is part of the TNF superfamily that binds to CD70 (also known as CD27L) expressed on activated T cells. Its agonism enhances T cell expansion and promotion of memory T cells, with a predilection toward IFN-γ production (96). It has gained interest as a target for immunotherapy, with ongoing phase I/II trials for anti-CD27 agonists such as varlilumab (97). With its promotion of IFN-γ production, initial studies postulated that CD27/CD70 agonism would worsen atherogenesis. Surprisingly, chronic inflammation from CD70 overexpression was atheroprotective, as the CD27/CD70 pathway has shown important roles in macrophage efferocytosis, oxLDL clearing, and promotion of atheroma T_reg survival (98, 99).

ICOS, or inducible costimulator, is a protein involved in CD4+ and CD8+ differentiation, survival and proliferation, promoting activation of cytotoxic T cells, enhancing effector T cell production of IL-4, IL-5, IL-10 and TNF-α, as well as promoting T_reg activity (100). Given its prominent role in the survival of both pro-tumor and anti-tumor T cells, both ICOS agonists and antagonists are in early phase cancer immunotherapy trials (100). In preclinical studies, ICOS inhibition was associated with aggravated atherosclerosis via a reduced T_reg population, as well as enhanced IFN-γ production and diminished IL-10 production (101, 102). Thus, contrary to CD40 and GITR, further study of CD27/CD70 and ICOS agonism may one day prove beneficial in atherosclerosis.

4. Investigational Co-Inhibitory Molecules: TIM-3 and LAG-3

TIM-3 is a negative regulatory immune checkpoint present on effector T cells, T_reg cells, and dendritic cells. It binds to four separate ligands, most notably to galectin-9 that triggers T_h1 and CD8+ T cell death (103, 104). Ongoing clinical trials are studying anti-TIM-3 antibodies such as cobolimab both as monotherapy or in combination with anti-PD-1 inhibitors(24).

Several preclinical studies have suggested that TIM-3 is a negative regulator of atherosclerosis. Foks et al. found that TIM-3 expression was increased in mice fed with atherogenic diet, and that TIM-3 antibody blockade was associated with larger atherosclerotic plaque size, increased lesion macrophage content, and decreased lesion T_reg cells (105). In atherosclerotic lesions, CD8+ T cells exhibits co-expression of PD-1 and TIM-3, and inhibiting both immune checkpoints was associated with increase TNF-α and IFN-γ and decrease in IL-10 and IL-4) when compared to singular PD-1 or TIM-3 blockade (106). Thus, the development of anti-TIM-3 checkpoint inhibitors pose theoretical risks of exacerbating atherosclerosis.

LAG-3 is a cell surface protein analogous to CD4 present on T cells and dendritic cells that binds MHC class II molecules and serves as an inhibitory signal. It directly competes with TCR and CD4 binding to MHC class II, suppresses T cell expansion and promotes CD4+ T cell differentiation into T_reg (107). LAG-3 targeting ICIs, such as relatlimab, are being investigated in Phase II and III clinical trials (108). One observational study of the Multiethnic Study of Atherosclerosis (MESA) cohort demonstrated that patients with CAD had higher LAG-3 levels, and LAG-3 was a significant coronary artery disease risk predictor (109). Despite this positive correlation, no causal relationship between LAG-3 and atherosclerosis has been established. Preclinical studies are needed to evaluate whether increased LAG-3 expression is a contributory or compensatory mechanism in atherosclerosis.

5. CD47-SIRPα Pathway

In addition to T cell mediated immune checkpoint inhibitors, novel classes of immune check therapies being explored include the macrophage-mediated immune checkpoint CD47. CD47 is an immunoglobulin-like molecule that binds to signal regulatory protein alpha (SIRPα) and impairs phagocytosis. In areas with high rates of apoptosis and cell turnover, such as within tumors and atherosclerotic necrotic cores, effective clearance of apoptotic cellular debris helps prevent further inflammatory response (110). This process, known as “efferocytosis” refers to programmed cell removal by which macrophages detect cell surface markers that signal phagocytosis, collectively termed “eat me” signals (111). By contrast, cells may express markers that impair phagocytosis, termed “don’t eat me” signals, such as CD47.

Kojima et al. demonstrated an upregulation of CD47 in both murine and human atherosclerotic plaque, particularly in the necrotic core (112). Treatment of atherosclerotic models with CD47 inhibitory antibodies markedly reduced atherosclerosis by restoring efferocytosis, evidenced by the enhanced removal of diseased vascular smooth muscles and macrophages in-vivo (112, 113). In addition, CD47 inhibition was shown to downregulate genes implicated in macrophage response to IL-1 and IFN-γ, leading to significant reduction in atherosclerotic inflammation in PET/CT imaging of mouse models (113). CD47 has also been suggested to reduce the ability of macrophages to remove opsonized targets, including the removal of opsonized clonal smooth muscle cells thought to give rise to the majority of cells in atherosclerotic plaques (114).

In recent years, CD47 inhibitor therapies have been used in clinical trials aimed at increasing tumor cell recognition and phagocytosis by macrophages. Magrolimab, the first-in-class anti-CD47 antibody, showed promising results in Phase 1B trials of relapsed and refractory non-Hodgkin’s lymphoma (115). It recently gained breakthrough therapy designation by the FDA, and ongoing trials are evaluating its efficacy in various hematologic and solid tumor malignancies. Interestingly, a small retrospective analysis on the non-Hodgkin’s lymphoma trial participants demonstrated a reduction of FDG uptake in the carotid arteries after 9 weeks of Magrolimab treatment, implying that CD47 inhibition reduces vascular inflammation (116). Thus, CD47 inhibition may be a shared pathway against both oncogenesis and atherogenesis.

Macrophage efferocytosis has also been shown to be influenced by T_reg cell activity. Proto et al. showed that T_reg cell expansion enhanced efferocytosis via expression of IL-13, which in turn increased IL-10 production in macrophages and led to apoptotic cell engulfment and clearance (117). Since existing PD-1 and CTLA-4 targeted ICI therapies have been suggested to suppress T_reg activity, decreased efferocytosis may be another avenue by which other classes of immune checkpoint therapies may exacerbate atherosclerosis. However, extensive research is needed to substantiate this potential link.

Cholesterol Metabolism and T Cell Mediated Immunity in Cancer

Clinically, obesity and metabolic syndrome have been clearly linked with increased cancer risk, mediated by many shared risk factors (118). One shared link is energy metabolism, where alterations in lipid metabolism can shape the immune system’s response to tumor activity. Cholesterol metabolism is essential for cancer progression, including the formation of cellular membranes during rapid proliferation and in tumor migration and invasion (119).

The features of metabolic syndrome, including hypertriglyceridemia, hyperglycemia and hypercholesterolemia, promotes a chronic inflammatory state that paradoxically blocks physiologic immune function (120). Chronic inflammation and hypercholesterolemia promote an increase in the production of myeloid-derived suppressor cells, or MDSCs, through a process termed “emergency myelopoiesis” (121). MDSCs inhibit T cell functions through alterations in TCR receptors that impair downstream signaling, secretion of TGF-β, IL-10 and cytokines that decrease effector T cell function, and overexpression of PD-L1 (122). In addition, Ma et al. utilized an Apoe^−/− mouse model traditionally used to study atherosclerosis to demonstrate negative regulatory effect of hypercholesterolemia on IL-9 levels, which impair CD8+ T cell differentiation and antitumor response (123). The enriched cholesterol content of the tumor microenvironment induces T cell exhaustion and the expression of immune checkpoints such as PD-1, TIM-3 and LAG-3 on CD8+ T cells (124).

Given the intricate relationship of hypercholesterolemia and repressed T cell function, it is anticipated that cholesterol levels modulate patient response to immunotherapy. Perrone et al. found that among patients receiving ICIs, baseline hypercholesterolemia was associated with improved overall survival rates (125). Other studies have established improved prognosis in obese patients treated with ICIs (126, 127). Thus it can be postulated that cancer patients with higher cholesterol levels may have worse T cell dysfunction by mechanisms described above, and subsequently exhibit a more pronounced response to the restoration of T cell function by immune checkpoint inhibition. However, these data are limited to date, and whether hypercholesterolemia is a causative factor in improved ICI response or simply a biomarker of chronic inflammation rendering patients more susceptible to immune restoration has yet to be established.

Cholesteryl esters are a storage form of cholesterol that are observed in increased numbers in the tumor microenvironment. One of the key enzymes in this process is ACAT1 (Acetyl-coenzyme A acetyltransferase), which promotes cholesterol esterification into the storage form and facilitates cholesterol transport in blood (128), has been studied in antitumor therapy and atherosclerosis. ACAT1 inhibition has been shown to decrease cancer cell migration and tumor progression in breast cancer and glioblastomas (129, 130). In the early stages of atherosclerosis, ACAT1 expression is increased in macrophages to enhance their ability to store free cholesterol, and the use of ACAT inhibitors had been shown to promote cell death (131). The ACAT1 inhibitor avasimibe has been studied in animals and humans with hyperlipidemia, however their effects on plasma cholesterol and atherosclerotic plaque size were inconsistent (132-134). Yang et al applied ACAT inhibition to cancer immunotherapy, and demonstrated that the addition of avasimibe to anti-PD-1 therapy was more effective in the treatment of melanoma in mice, with demonstrated enhancement in CD8+ T cell antitumor activity (135). Together, these findings imply an important relationship between hypercholesterolemia, a major driver of atherosclerosis, and tumor progression.

Clinical Implications of ICI Therapy on ASCVD

Recent clinical data have emerged to suggest an association between ICI use and accelerated atherosclerosis and atherosclerotic cardiovascular events. Initially, a case series of 11 patients suggested that PD-1 blockade may actually improve complicated plaque burden, as PD-1 blockade was associated with regression of atherosclerotic plaque in 3 patients (27%), no change in 7 (64%) and an increase in 1 (9%). However, this study was limited by a short CT scan follow up period and imaging protocols that were not intended for vascular imaging (136). Since, two case reports have linked PD-L1 and PD-1 inhibitors with rapid interval progression of CAD on left heart catheterization and fatal ACS in metastatic lung and giant bone cell cancer, respectively (137, 138).

Several single center studies have documented the incidence of ASCVD in ICI therapy. In a retrospective analysis of ICI clinical trials where PD-1 and PD-L1 inhibitors including nivolumab, pembrolizumab, atezolizumab, avelumab, and durvalumab were tested in patients with NSCLC, Hu et al. reported a 1% incidence of myocardial infarction and 2% incidence of stroke (139). In a single center registry by Oren et al. of 3,326 patients with any malignancy on ICI therapy including atezolizumab, avelumab, ipilimumab, nivolumab and pembrolizumab, there was a 7% incidence for both myocardial infarction and stroke within a 16 month period (140).

In the largest single center study to date, Drobni et al. compared 2842 patients on any ICIs (the majority of which being PD-1 inhibitors) with controls matched for age, cancer type and cardiovascular history on their risk of ASCVD related events over a 2 year follow up period. ICI use was associated with >4-fold greater risk for composite cardiovascular events (HR 4.7, [95% CI 3.5-6.2, p<0.001), >7 fold greater risk of myocardial infarction (HR 7.2 [95% CI 4.5-11.5, p<0.001], 3-fold greater risk of coronary revascularization (HR 3.3, [95% CI 2.0-5.5], p<0.001) and 4-fold greater risk of stroke (HR 4.6 [95% CI 2.9-7.2], p<0.001) (7). Small scale human imaging and histologic studies have attempted to substantiate the possibility that ICI therapy increases atheromatous inflammation and atherosclerosis. In Calabretta et al., PET scans from twenty melanoma patients treated with ICIs (80% PD-1 inhibitors, 5% CTLA-4 inhibitors and 15% combination) showed significant elevation in FDG uptake of all aortic segments about 6 months post-treatment, particularly in noncalcified and mildly calcified segments (p<0.001) (141). However, in a study by Poels et al., ten melanoma patients treated with pembrolizumab or combination nivolumab/ipilimumab showed no absolute change in overall vascular FDG-positive inflammation after 6 weeks.(142) These discrepant results may be due to small sample sizes and varying follow up periods. In a sub-study by Drobni et al., 40 melanoma patients were evaluated via CT at three different time points, and ICI treatment was shown to accelerate the rate of atherosclerotic plaque progression 3-fold (from 2.1%/yr pre-ICI to 6.7%/yr after ICI, p=0.02). In an autopsy study of 11 patients on ICI therapy, although the absolute number of T cells was unchanged, there was an increased ratio of CD3+ cells to CD68+ cells leading to a shift from macrophage- to lymphocyte-predominant inflammation (143). Future large-scale, long term imaging and histologic studies will be needed to further clarify the role of ICI on atheromatous plaque progression and immune response.

Role of Pharmacotherapy in Cardiovascular Risk Reduction in Patients Receiving ICIs

Growing clinical data signaling increased atherosclerosis in ICI therapy invite consideration of pharmacotherapies to reduce cardiovascular events in this patient population. In addition to lipid lowering, statins have been associated with plaque stabilization, endothelial dysfunction reversal, and inflammation reduction. How statins reduce inflammation are not fully understood, but may involve inhibition of beta-2 integrin leukocyte function antigen-1 (LFA-1), an adhesion molecule with a role in T cell activation (144). In Drobni et al. the markedly increased rate of aortic plaque progression associated with ICI therapy was attenuated by the use of statins, though no direct comparisons of cardiovascular events based on statin use were performed (7). PCSK9 inhibitors are an increasingly popular class of monoclonal antibodies that reduce serum LDL and atherosclerotic events in higher risk patients (145). In contrast to statins, there is much less known about potential anti-inflammatory effects of PCSK9 inhibitors.

The potential for cholesterol levels to modulate ICI response raises important considerations for the effects of these cholesterol lowering agents on ICI efficacy. Despite the correlation between hypercholesterolemia and improved outcomes of ICI therapy, statins and PCSK9 inhibitors have shown preliminary evidence of synergistic benefit when paired with ICI therapy independent of their cholesterol lowering effects. In particular, statins have been shown to inhibit protein prenylation leading to enhanced antigen presentation that may be synergistic with immunotherapy (146). Several clinical studies in patients with advanced NSCLC and malignant pleural mesothelioma treated with ICIs demonstrated that statins were associated with increased response rate, improved time to treatment failure, progression free and overall survival (147, 148). Liu et al. showed that proprotein convertase substilisin/kexin type 9 (PCSK9) inhibition with evolocumab synergized with anti-PD-1 therapy to suppress tumor growth by increasing the expression of MHC I proteins and enhancing lymphocyte proliferation into the tumor (149). These findings highlight the complex interplay between cholesterol metabolism and the immune system that requires further research, but demonstrate that statins and PCSK9 inhibitors have the potential to both enhance ICI efficacy and treat ICI related atherosclerosis. However, Drobni et al. recently demonstrated increased risk of myopathy in patients treated concurrently with statins and ICIs (150). Thus, further study is needed to confirm the safety of existing therapies and identify novel therapeutics in the treatment of ASCVD in the context of ICI use.

Conclusions

Numerous studies have demonstrated the role of various immune checkpoint pathways in atherogenesis, and emerging clinical studies have begun describing an association between ICI use and increased risk of ASCVD. Taken together, these reports indicate that the cardiovascular effects of immune checkpoint therapies extend beyond the rare incidences of myocarditis. As ASCVD is one of the most prominent causes of morbidity and mortality worldwide, if further clinical studies confirm this association, the risk of major adverse cardiovascular events in patients receiving ICIs must be carefully considered.

The current level evidence linking ICI therapy to atherogenesis and atherosclerotic events is not without its limitations. Preclinical studies thus far have not clearly delineated how immune checkpoint alteration will impact each stage of atherosclerosis. It has become increasingly recognized that the microenvironmental context of cell death and apoptosis is important in determining whether the net effect is atherogenic or atheroprotective (111). Thus, the effect of ICIs on atherosclerosis may prove to be more heterogeneous and evolve as clinical disease progresses. Careful mechanistic understanding of the effect of ICI at each stage of atherosclerosis is needed to determine the timing and need of interventions.

On a clinical level, the linkage between ICI use and atherosclerosis has only been suggested by smaller observational studies, and extensive further evidence with larger sample sizes and longer study duration are needed to confirm this association and estimate the rate of adverse cardiovascular events in this population. If future studies continue to demonstrate significant ASCVD burden in patients receiving ICIs, effort must be taken to identify atherosclerotic development and control reversible atherosclerosis risk factors. In this effort, integration of baseline and routine atherosclerotic imaging in these patients may help to quantify atherosclerotic burden.

Patients who develop ASCVD from ICI therapy represent a unique patient population that warrant our investigation of potential pharmacotherapeutic options. The role of statins, PCSK9 inhibitors and other pharmacotherapies should be explored in animal models, and randomized controlled trials are needed to prove their effectiveness and safety in this population.

Highlights.

• Immune checkpoint therapy may increase atherosclerotic cardiovascular disease (ASCVD).

• Cholesterol metabolism modulates the response to immune checkpoint therapies and may be a common target for treatment of both cancer and atherosclerosis.

• Further research should define cardiovascular endpoints for patients receiving immune checkpoint therapies and standardize treatment and surveillance strategies.

Nonstandard Abbreviations

ASCVD

Atherosclerotic cardiovascular disease

CANTOS

Canakinumab Anti-Inflammatory Thrombosis Outcomes Study

COLCOT

Colchicine Cardiovascular Outcomes Trial

CTLA-4

Cytotoxic T-lymphocyte associated protein 4

ICI

Immune checkpoint inhibitor

IFN-γ

Interferon gamma

LAG-3

Lymphocyte-activation gene 3

LoDoCo

Low-Dose Colchicine

PD-1

Programmed cell death protein 1

PD-L

Programmed cell death ligand