A narrative review of immune checkpoint inhibitor-related cardiotoxicity in lung cancer: mechanisms, risk stratification, and clinical management

Abstract

Background and Objective

Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of advanced lung cancer. However, their associated cardiac toxicity constitutes a life-threatening immune-related adverse reaction, with myocarditis being the most severe form, carrying a mortality rate of nearly 40%. Patients with lung cancer face a distinctively higher risk of ICI-related cardiotoxicity compared with other malignancies, driven by lung cancer-specific factors including tumor microenvironment hypoxia and fibrosis, prior thoracic radiotherapy, smoking-related comorbidities, and potential antigen cross-reactivity. Given the unique high-risk profile of ICI-related cardiac injury in lung cancer patients, this review integrates multidisciplinary clinical evidence and translational research findings to provide practical and forward-looking strategies for the standardized management of ICI-related cardiac toxicity. Furthermore, this study emphasizes risk stratification and the exploration of underlying mechanisms, proposing feasible research directions to promote the optimization of the balance between efficacy and safety in lung cancer immunotherapy, thereby providing a sound reference for the further development of cardiac oncology practice in lung cancer.

Methods

A narrative review was conducted using literature retrieved from PubMed, ClinicalTrials.gov, the Chinese Clinical Trial Registry (ChiCTR), and Google Scholar, covering the period from January 2000 to March 2026. Search topics focused on ICI related targets, various lung cancer subtypes, and associated cardiovascular and immune-related adverse events (irAEs). Original research articles, reviews, clinical trials, and case reports focusing on ICI-related cardiac toxicity in patients with lung cancer were included. To ensure comprehensiveness and minimize omissions, relevant abstracts from the 2020 to 2026 meetings of the American Society of Clinical Oncology (ASCO), the European Society for Medical Oncology (ESMO) and the World Conference on Lung Cancer (WCLC) were manually screened, and reference lists of included studies were cross-checked to identify additional eligible literature.

Key Content and Findings

This review synthesizes current evidence on the epidemiology, clinical features, and mechanisms of ICI-related cardiotoxicity specifically in lung cancer, with an emphasis on how lung cancer-specific pathophysiological features amplify susceptibility to immune-mediated cardiac injury. We summarize key biomarkers and predictive models for early detection and risk stratification and propose a practical clinical management framework tailored to this high-risk population.

Conclusions

Although ICI immunotherapy has significantly improved survival outcomes in patients with advanced lung cancer, ICI-related cardiovascular toxicity remains a critical and life-threatening clinical challenge. Optimizing biomarker monitoring, accurate risk stratification and standardized cardiological management are essential for reducing adverse cardiac events and improving the long-term prognosis of lung cancer patients receiving immunotherapy.

Introduction

Background

Lung cancer remains the leading cause of cancer-related mortality worldwide, with an estimated 2.48 million new cases and 1.82 million deaths reported in 2022, underscoring its persistent global health burden (1-3). Conventional treatments, including surgery, chemotherapy, and radiotherapy, offer limited benefit in advanced disease stages. The advent of immune checkpoint inhibitors (ICIs) targeting programmed death receptor 1 (PD-1), programmed death-ligand 1 (PD-L1) and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) has revolutionized the treatment landscape for lung cancer, significantly improving patient survival rates (4,5). However, these drugs also present safety concerns that cannot be overlooked, namely immune-related adverse events (irAEs). Among the various adverse reactions, cardiotoxicity, although relatively rare, progresses extremely rapidly and has a high mortality rate; it is particularly prominent in lung cancer patients, posing a serious challenge to immunotherapy (6,7).

Rationale and knowledge gap

Multiple meta‑analyses demonstrating that lung cancer is associated with a significantly elevated risk of ICIrelated cardiotoxicity compared with other solid tumors (8-10). In a large cohort study involving 54,409 patients with solid tumors, the incidence of ICI-related cardiovascular immune-related adverse events (C-ICIAE) was 8%. Lung cancer accounted for the highest proportion of these cases at 59% (11). However, it should be noted that the 59% estimate derives from a single real‑world study and requires further validation in larger, prospective cohorts.

Another multicenter real-world study involving 23 million lung cancer patients also demonstrated that the incidence of cardiovascular adverse events within 180 days of ICI treatment was 4.0% (12). Taking all this evidence together, it is clear that lung cancer patients are already a high-risk group for ICI-related cardiotoxicity, rather than this being caused solely by exposure to the treatment. This disparity is likely linked to the tumor’s intrinsic biological characteristics and unique tumor microenvironment (TME). Tumor development is driven by genetic mutations and abnormal signaling pathways, which not only determine the tumor’s biological behavior but also influence the body’s systemic immune response through complex interactions within the TME.

The interplay between the tumor’s intrinsic characteristics and the TME jointly determines immunotherapy efficacy and the occurrence of off-target immunotoxicity. It is worth noting that lung cancer is characterized by a high mutational burden and is often associated with smoking-related chronic inflammation, prior vascular damage and chest radiotherapy history. Against this backdrop, combined with unique lung TME remodeling, systemic immune activation is likely to be further amplified, thereby increasing patients’ susceptibility to immune-mediated cardiac injury. However, most of the current evidence stems from pan-cancer analyses, and data specifically targeting lung cancer remain scarce, highlighting a critical gap in translational research. For this reason, there is an urgent need to establish a research framework focused on lung cancer.

Objective

To address this research gap, this review synthesizes available clinical and translational evidence, focusing on the mechanisms of action, risk stratification and management strategies of ICI-related cardiotoxicity in lung cancer. This work aims to optimize risk assessment for high-risk populations and facilitate more precise and personalized clinical decision-making. We present this article in accordance with the Narrative Review reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1372/rc).

Methods

A narrative review was conducted using literature retrieved from PubMed, ClinicalTrials.gov, the Chinese Clinical Trial Registry (ChiCTR), and Google Scholar, covering publications issued from January 2000 to March 2026. All search terms were combined using Boolean operators (AND/OR) and categorized into three core clusters: ICI-related targets, various lung cancer subtypes, and cardiovascular and irAEs.

This search was limited to relevant literature on human subjects published in English within the specified time frame. Eligible publication types included original research articles, review articles, randomized controlled trials, retrospective cohort studies, meta-analyses, clinical trials, case reports, and clinical guidelines, all focusing on ICI therapy-associated cardiac toxicity in patients with lung cancer. To ensure comprehensiveness and minimize omissions, relevant abstracts from the 2020 to 2026 meetings of the American Society of Clinical Oncology (ASCO), the European Society for Medical Oncology (ESMO) and the World Conference on Lung Cancer (WCLC) were manually screened, and reference lists of included studies were cross-checked to identify additional eligible literature. The complete search strategy is summarized in Table 1.

Table 1. Search strategy summary.

Item	Specification
Date of search	March 31, 2026
Databases and other sources searched	PubMed, ClinicalTrials.gov, ChiCTR, Google Scholar, ASCO, ESMO, WCLC meeting
Search terms used	Immune checkpoint inhibitor-related targets: “PD-1”, “PD-L1”, “CTLA-4”, “LAG-3”, “TIGIT”, “TIM-3”, “VISTA”, “BTLA”, “CD47”; lung cancer subtypes: “non-small cell lung cancer”, “small cell lung cancer”, “lung squamous cell carcinoma”, “lung adenocarcinoma”; cardiovascular and immune-related adverse events: “myocarditis”, “pericarditis”, “pericardial effusion”, “cardiomyopathy”, “arrhythmia”, “heart failure”, “cardiac arrest”, “myocardial infarction”, “vasculitis”, “thrombosis”, “hypertension”
Timeframe	January 1, 2000 to March 30, 2026
Inclusion and exclusion criteria	Inclusion: English peer-reviewed articles, original studies, reviews, clinical trials and case reports focusing on ICI-related cardiac toxicity in lung cancer patients. Exclusion: non-English literature, letters, editorials, and preclinical animal studies
Selection process	Study selection conducted by Y.S. and X.W.; discrepancies resolved by discussion and consensus

ASCO, American Society of Clinical Oncology; BTLA, B- and T-lymphocyte attenuator; CD47, cluster of differentiation 47; ChiCTR, Chinese Clinical Trial Registry; CTLA-4, cytotoxic T-lymphocyte-associated antigen 4; ESMO, European Society for Medical Oncology; ICI, immune checkpoint inhibitor; LAG-3, lymphocyte activation gene 3; PD-1, programmed death receptor 1; PD-L1, programmed death-ligand 1; TIGIT, T cell immunoreceptor with Ig and ITIM domains; TIM-3, T-cell immunoglobulin and mucin-domain containing-3; VISTA, V-domain Ig suppressor of T-cell activation; WCLC, World Conference on Lung Cancer.

Epidemiological characteristics and risk factors of ICIs-related cardiac toxicity in patients with lung cancer

ICI-related myocarditis (ICI-M) is rare but highly lethal, with mortality approaching 40% (8). In patients with lung cancer, accumulating clinical evidence indicates that ICI monotherapy or ICI combined with chemotherapy significantly increases the risk of cardiotoxicity compared with conventional chemotherapy, whereas dual ICI combination therapy does not further elevate this risk relative to ICI monotherapy (9,10). A recent umbrella review synthesizing data from multiple systematic reviews and meta-analyses further quantified that non-small cell lung cancer (NSCLC) patients receiving combination ICI-based therapies face a 97% higher risk of all-grade cardiotoxicity compared with ICI monotherapy, with ICI-induced myocarditis carrying a mortality rate of 37.7% (13). A prospective study reported a cumulative incidence of cardiac toxicity of 19.4% within 3 months of ICI initiation (14). Consistently, a nationwide multicenter retrospective analysis of 55,219 patients with advanced NSCLC identified myocarditis in 0.29% of ICI-treated patients, with a median onset of 59 days and 64% of cases occurring within the first 3 months (15). During a median follow-up of 6 months, the overall incidence of cardiovascular adverse events reached 13.3% (16).

Cardiotoxicity risk also varies substantially across immune checkpoint regimens. CTLA-4-based combinations with chemotherapy and PD-L1-targeted therapies appear relatively safer, whereas PD-1 monotherapy confers a higher risk than PD-L1 blockade, and combined PD-1 plus CTLA-4 therapy is associated with the greatest incidence of cardiac adverse events (17). Emerging evidence suggests that anti-lymphocyte activation gene 3 (LAG-3) therapy may induce autoimmune activation and myocardial involvement (18,19); however, both preclinical models and NSCLC cohorts report lower cardiac toxicity compared with anti-PD-1 therapy (20,21). In contrast, the cardiotoxic mechanisms associated with T cell immunoreceptor with Ig and ITIM domains (TIGIT) and V-domain Ig suppressor of T-cell activation (VISTA) inhibition remain poorly defined. Collectively, these findings underscore the need for improved risk stratification and early cardiac monitoring to mitigate cardiotoxicity during lung cancer immunotherapy.

Clinical manifestations of ICI-associated cardiac toxicity in lung cancer

Common manifestations of cardiac toxicity associated with ICIs in lung cancer include myocarditis, pericarditis, arrhythmias, and vasculitis (22,23), which have complex and diverse symptoms.

Myocarditis

ICI-M is the most frequent and lethal cardiac toxicity in lung cancer, despite its low incidence (0.2–4.6%) (24,25). Clinical presentations vary considerably among patients, with symptoms including chest pain, dyspnea, palpitations, syncope and fatigue. It is worth noting that ICI-M has an early onset, with a median time to onset of 27 days; 76% of cases occur within the first 6 weeks of treatment, and the case-fatality rate can be as high as 46% (26). Although clinical presentations are diverse, and electrocardiogram (ECG) abnormalities and elevated biomarkers are relatively common, not all patients will exhibit these features (27-29). Crucially, early ECG changes such as sinus tachycardia and prolonged corrected QT intervals (QTc) often indicate the potential for rapid deterioration and warrant high vigilance (30). Once biomarkers show a significant increase, cardiac magnetic resonance imaging (MRI) or endomyocardial biopsy should be performed immediately to confirm the diagnosis y (31,32).

Pericardial diseases

Pericardial diseases are a common manifestation of ICI-related cardiac toxicity in lung cancer patients, including pericarditis, pericardial effusion and cardiac tamponade. They account for 56.3% of ICI-related cardiac adverse events, with a median time to onset of 30 days after treatment initiation (33). Clinical manifestations range from dyspnea, chest pain, and tachycardia to asymptomatic presentations, in which echocardiography is essential for detection and guidance of pericardiocentesis (34). As etiological differentiation among tumor progression, immune-mediated pericarditis, and infection is often challenging, pericardiocentesis combined with pericardial biopsy provides critical diagnostic evidence (34).

Arrhythmias and conduction abnormalities

Arrhythmias and conduction abnormalities are relatively common adverse cardiac events during ICI treatment for lung cancer. Among these events, atrial fibrillation accounts for approximately 30%, ventricular arrhythmias for 27%, and atrioventricular block for 17% (35). Conduction abnormalities are strongly associated with cardiovascular mortality, with reported rates of up to 80% compared with 16% in patients without conduction disorders (36). Most patients are asymptomatic or present with mild symptoms, while electrocardiography may reveal PR prolongation or bundle branch block (36). Histopathological evidence of lymphocytic infiltration in the sinoatrial and atrioventricular nodes supports a T cell-mediated mechanism of conduction system injury (37). These findings underscore the importance of baseline ECG assessment and 24-hour Holter monitoring prior to ICI initiation.

Vasculitis

Vasculitis complications under ICI therapy are rare (<1%) (38) but clinically consequential, predominantly affecting large vessels such as in giant cell arteritis, aortitis, and primary angiitis of the central nervous system (39,40). Symptomatology varies with vascular territory, with a median onset of 55 days (33). ICIs activate adventitial dendritic cells (DCs), induce CD4⁺ T cell polarization and proinflammatory cytokine secretion, recruit CD8⁺ T cells and monocytes, and trigger vascular remodeling, intimal hyperplasia, and occlusion, leading to vasculitis (41). Notably, this proinflammatory milieu may also accelerate atherosclerosis, increasing the risk of major cardiovascular events such as myocardial infarction and ischemic stroke (42).

Although uncommon, additional cardiovascular complications beyond the above cardiac toxicity manifestations are associated with ICIs, including thromboembolism, myocardial infarction, heart failure, valvular disease, hypertension, and others. Early baseline cardiovascular assessment and dynamic monitoring during treatment remain essential to mitigate potential risks and optimize patient outcomes.

Pathogenesis of ICI-induced cardiotoxicity in lung cancer

ICIs restore antitumor immunity by blocking inhibitory pathways such as CTLA-4 and PD-1/PD-L1, thereby releasing tumor-mediated T-cell suppression (43,44). However, PD-1/PD-L1 blockade may also enhance myocardial-reactive T-cell activity, disrupt the cardioprotective function of PD-L1 in cardiomyocytes, and upregulate proinflammatory mediators such as growth arrest and DNA damage-inducible-gene 153 (GADD153), ultimately promoting cardiac injury (45,46).

Lung cancer-specific TME as a mechanistic amplifier of ICI-related cardiotoxicity

Within the lung cancer TME, chronic exposure to smoking-related carcinogens and environmental insults drives oncogenic alterations (e.g., KRAS and TP53 mutations), accompanied by persistent hypoxia and extensive immune cell remodeling. This results in a highly dysregulated immune milieu characterized not only by immunosuppression but also by a state of latent inflammatory priming (47-56).

Under physiological conditions, this preconditioned environment remains relatively constrained; however, upon ICI therapy, the abrupt release of immune checkpoint inhibition may trigger exaggerated systemic immune activation. Reactivated T cells and pro-inflammatory cytokines can extend beyond the tumor site and target off-tumor tissues, including the myocardium, leading to immune-mediated cardiac injury.

Importantly, hypoxia-associated signaling and profibrotic pathways further amplify systemic inflammation and endothelial dysfunction, thereby lowering the threshold for cardiac injury. Collectively, these lung cancer-specific microenvironmental features may act as a mechanistic amplifier, predisposing patients to a higher incidence and greater severity of ICI-related cardiotoxicity compared with other malignancies (Figure 1).

Figure 1.

Lung cancer-specific tumor microenvironment amplifies susceptibility to ICI-related cardiotoxicity. In lung cancer, a hypoxic and immunologically dysregulated tumor microenvironment primes systemic immune activation. Upon ICI therapy, this preconditioned milieu triggers exaggerated T-cell-mediated inflammation and myocardial injury, while hypoxia and fibrotic signaling further amplify susceptibility to ICI-related cardiotoxicity. APC, antigen-presenting cell; CAFs, cancer-associated fibroblasts; HIF-1α, hypoxia-inducible factor-1α; ICI, immune checkpoint inhibitor; MHC II, major histocompatibility complex class II; MPS, macrophage phenotype switching; PIF, pulmonary interstitial fibrosis; TAM, tumor-associated macrophages; TCR, T cell receptor.

Thus, the TME in lung cancer should be viewed not only as a regulator of antitumor immunity but also as a critical determinant of systemic irAEs, providing a mechanistic basis for the observed cardiotoxicity risk in this population. This perspective may partially explain why cardiotoxicity associated with ICIs appears more frequent and severe in lung cancer than in several other solid tumors.

Underlying immune mechanisms of ICI cardiotoxicity

In lung cancer, CTLA-4 inhibition depletes regulatory T cells within the TME and promotes Th17 differentiation with increased IL-17A expression, triggering inflammatory cascades via the NOD-like receptor family pyrin domain-containing 3 (NLRP3)/IL-1β and myeloid differentiation primary response 88 (MyD88) pathways (57-59). Although ICIs have demonstrated substantial efficacy in NSCLC, excessive immune activation can trigger autoimmune responses and cardiotoxic irAEs (60), involving T cells, macrophages, exosomes, and other inflammatory mediators (Figure 2).

Figure 2.

ICI-induced immune-mediated myocardial injury in lung cancer: from T-cell activation to cardiac toxicity. ICIs block inhibitory signals to activate T cells and kill tumor cells. Overactivated T cells may damage cardiomyocytes via multiple mechanisms, while smoking, pollution, POL and TRT further raise the risk of adverse cardiac events. ACR, antigen cross-recognition; AMA, antibody-mediated autoimmunity; AS, atherosclerosis; C-ICIAE, cardiac ICI-related adverse events; CAA, coronary artery atherosclerosis; CRD, cardiac radiation dose; CUTM, commonly used treatment methods; ICB & CII, immune checkpoint blockade and cardiac immune imbalance; ICI, immune checkpoint inhibitor; ICM, inflammation and cytokine-driven immune injury mechanisms; IPF, idiopathic pulmonary fibrosis; OMH, other mechanisms hypothesis; POL, pulmonary space-occupying lesion; TRT, thoracic radiotherapy.

Antigen cross-recognition

In ICI-M, activated T cells indiscriminately attack both myocardial and tumor tissues due to the “molecular mimicry” between myocardial and tumor antigens. Histopathological examination of myocardial tissue from such patients revealed T-cell infiltration resembling that seen in transplant rejection (61). As a myocardial-specific structural protein, α-myosin is considered a key autoantigen in ICI-M and a primary target of T-cell misrecognition (62). Thus, the cardiotoxicity of ICIs in lung cancer likely stems from T-cell “cross-reactivity” triggered by structural similarities between tumor and myocardial antigens. However, most of the evidence regarding α-myosin as a key autoantigen derives from murine models and small case series, and large-scale validation in lung cancerspecific cohorts is lacking.

Immune checkpoint blockade and cardiac immune imbalance

Immune checkpoints (PD-1, PD-L1, CTLA-4) act as cardiac “immune gatekeepers” restraining T-cell activity, limiting cytokine release, and preserving myocardial tolerance (63,64). ICIs disrupt these pathways in cancer therapy, unleashing aberrant T-cell activation that triggers myocarditis and pericarditis (65). Preclinical models, including PD-1/PD-L1-deficient and Pdcd1^−/−Ctla4^+/− mice, show enhanced T-cell infiltration, cytokine elevation, conduction defects, and high mortality, highlighting the essential role of intact PD-1/PD-L1 and CTLA-4 signaling in preventing ICI-induced myocarditis (66-71). It should be noted that much of this mechanistic evidence comes from preclinical Pdcd1⁻/⁻Ctla4⁺/⁻ mouse models, and translation to human ICI-M requires further clinical confirmation.

Antibody-mediated autoimmunity

During lung cancer immunotherapy, ICIs disrupt immune tolerance by inducing autoantibody production, activating immune cells, and amplifying inflammatory cytokine release, thereby triggering irAEs. Evidence indicates that patients with myocarditis and inflammatory dilated cardiomyopathy have cardiac-specific autoantibodies and T cells targeting the α-myosin heavy chain (62,72,73). In PD-1 deficient mice, anti-myosin autoantibodies provoke fulminant myocarditis with marked CD4⁺ and CD8⁺ T-cell infiltration restricted to cardiac tissue (74).

High-titer autoantibodies against myosin and cardiac troponin I further drive Ca²⁺ overload and myocardial dysfunction (75). Clinically, when lung cancer patients receive treatment with PD-1 inhibitors, cardiac antigens previously exposed to radiotherapy may be reactivated and exposed, thereby inducing autoantibody-mediated CD8⁺ T-cell infiltration and accumulation, ultimately leading to cardiac toxicity (76).

Such autoantibodies hold promise as early biomarkers, supporting the identification of high-risk patients through dynamic monitoring and thereby enabling early intervention (77). These findings are primarily derived from PD-1-deficient murine models and small clinical case series; larger prospective studies are needed to validate the clinical utility of autoantibodies as early biomarkers.

Inflammation and cytokine-driven immune injury mechanisms

Emerging clinical and translational evidence suggests that cytokine dysregulation plays a significant role in irAEs associated with ICI therapy in lung cancer patients. Patients with immune-related myocarditis exhibit significantly elevated levels of creatine kinase, myoglobin, interleukin-6 (IL-6) and IL-10, a finding that strongly suggests a close association between inflammatory mediators and severe cardiac toxicity (78,79). Longitudinal analyses reveal that irAEs are preceded by low baseline levels of C-X-C motif chemokine ligand 9 (CXCL9), CXCL10, CXCL11, and CXCL13, followed by rapid post-treatment upregulation (80). In advanced NSCLC, increased granulocyte colony-stimulating factor (G-CSF) and Regulated on Activation, Normal T Cell Expressed and Secreted (RANTES) and reduced leptin further distinguish irAE-prone patients (81). Notably, post-treatment increases in IL-6, CXCL5, CXCL9, and CXCL10, together with interferon-gamma (IFN-γ) suppression, predict a heightened risk of irAEs and cardiotoxicity (82,83). Collectively, cytokine dynamics offer predictive value and suggest inflammatory targeting as a potential preventive strategy. The cytokine data discussed here are largely derived from small observational cohorts and require validation in larger, prospective studies with standardized sampling protocols.

ICI-related atherosclerosis

Beyond acute myocarditis, ICI therapy has been shown to accelerate atherosclerosis progression, representing a chronic and increasingly recognized form of ICI-induced cardiovascular toxicity. Drobni et al. demonstrated that ICI treatment was associated with a three-fold increase in atherosclerotic plaque volume on serial imaging (42).

Mechanistically, PD-1 signaling exerts anti-atherosclerotic effects by suppressing Th1 cytokines and promoting regulatory T cells; PD-1 blockade disrupts this protective pathway, thus elevating the incidence of critical cardiovascular disorders involving cardiac infarction and stroke. Zheng et al. recently pointed out that ICI-induced cardiovascular toxicities primarily manifest as acute myocarditis and chronic atherosclerosis, both driven by excessive immune activation; in particular, phenotypic remodeling of T cells and macrophages constitutes a key factor in the underlying pathogenesis (84). Han et al. comprehensively reviewed the pathophysiological mechanisms by which ICIs promote atherosclerosis and discussed risk assessment and intervention strategies (85).

This mechanism is particularly relevant to lung cancer patients, given their high prevalence of smoking-related vascular comorbidities and pre-existing cardiovascular disease. The evidence linking ICI therapy to accelerated atherosclerosis is primarily derived from serial imaging studies with modest sample sizes; larger longitudinal studies are needed to confirm these findings.

Other mechanisms hypothesis (OMH)

The OMH encompasses multiple external and host-related factors that modulate the risk and severity of ICI-related cardiotoxicity.

Several additional factors may modulate the risk of ICI-related cardiotoxicity in lung cancer patients. Pre-existing autoimmune diseases, affecting up to 14–25% of patients with NSCLC, share mechanistic overlap with irAEs through dysregulation of PD-1/PD-L1 and CTLA-4 pathways (86-90). ICI-induced gut microbiota dysbiosis may further amplify systemic immune activation (91,92). However, the most clinically relevant amplifying factors for lung cancer patients are underlying cardiovascular burden, chronic hypoxia due to smoking-related lung disease, prior thoracic radiotherapy, and prior treatments with cardiotoxic effects; the combined effect of these factors further exacerbates immune-mediated cardiac injury (93-102). The modulating effects of pre-existing autoimmune diseases, gut microbiota dysbiosis, and prior thoracic radiotherapy on ICI-related cardiotoxicity are largely supported by observational data and hypothesis-generating studies; mechanistic confirmation and prospective validation are warranted.

Clinical studies on cardiotoxicity associated with ICIs in lung cancer treatment

Table 2 summarizes 19 clinical trials (21,24,38,103-118), and Table 3 summarizes seven retrospective clinical studies (16,25,119-123), with a combined total of 26 studies, all focused on ICI-related cardiotoxicity in lung cancer treatment. These studies included 43,747 patients with follow-up durations ranging from 5.5 to 72 months and provided multidimensional insights into the characteristics and clinical implications of this cardiotoxicity. Specifically, when PD‑1 inhibitors are used as monotherapy, the incidence of lung cancer‑related myocarditis ranges from 0.2% to 0.86%; when combined with CTLA‑4 inhibitors, the incidence rate increases to 1.1%; and the incidence of myocarditis can be as high as 8% with combination immunotherapy plus radiotherapy. The cardiac toxicity of PD-L1 inhibitors is significantly characterized by a 3% incidence of vasculitis, reflecting the specificity of different target ICIs in toxicity expression.

Table 2. Clinical trials of ICI-related cardiovascular toxicity in lung cancer treatment.

ICIs	Target	Lung cancer type/stage	National clinical trial number	Total patients, n	ICI-treated patients, n	Follow-up (months)	ICI-related cardiovascular irAEs (%)	References
Pembrolizumab	PD-1	Locally advanced/metastatic NSCLC	NCT02220894	1,274	637	12.8	0.2% myocarditis; 0.2% acute heart failure	(24)
Cemiplimab		Locally advanced/metastatic NSCLC	NCT03088540	712	357	35	0.3% myocarditis; 0.3% heart failure; 0.3% cardiopulmonary arrest	(103)
Pembrolizumab		Metastatic non-squamous NSCLC	NCT02578680	616	410	60	0.2% myocarditis; 0.5% vasculitis	(104)
Pembrolizumab		Advanced NSCLC	NCT01905657	1,034	346	13.1	0.29% heart attack	(105)
Sugemalimab		IV NSCLC	NCT03789604	479	320	17.8	0.3% myocarditis; 0.9% heart failure; 1.3% hypertension	(106)
Pembrolizumab		IB–IIIA NSCLC	NCT02504372	1,177	590	35.6	0.86% myocarditis; 0.17% vasculitis; 6% hypertension	(38)
Pembrolizumab		IV squamous cell NSCLC	NCT02775435	559	278	60	0.7% vasculitis	(107)
Sintilimab		IIIB–IVB NSCLC	NCT04167657	14	14	40.1	7% myocarditis (+ thoracic radiotherapy)	(108)
Atezolizumab	PD-L1	IV NSCLC	NCT02409342	572	285	17	3% vasculitis	(109)
Adebrelimab		ES-SCLC	NCT03711305	462	230	13.5	2.6% hypertension	(110)
Avelumab		IIIB/IV recurrent NSCLC	NCT02395172	792	396	18.9	0.3% myocarditis	(111)
Atezolizumab		Recurrent/metastatic NSCLC	NCT03563716	135	68	30.4	1.5% myocarditis; 1.5% cardiomyopathy; 1.5% mitral valve incompetence	(112)
Durvalumab		III NSCLC	NCT02125461	709	473	14.5	0.2% cardiomyopathy; 0.2% cardiopulmonary failure; 0.2% myocardial infarction; 0.2% aortic dissection; 0.2% heart failure; 0.4% cardiac arrest	(113)
Nivolumab, ipilimumab	PD-1, CTLA-4	Late-stage squamous cell NSCLC	S1400 (Lung-MAP sub-study)	252	125	29.5	0.8% myocarditis; 0.8% heart failure; 0.8% stroke; 0.8% thromboembolism; 0.8% cardiac arrest; 3.2% hypertension	(114)
Pembrolizumab, ipilimumab		IV NSCLC	NCT03302234	568	284	24	1.1% myocarditis	(115)
Nivolumab, ipilimumab		Limited/extensive stage SCLC	NCT01928394	216	61	10.1	1.6% cardiomyopathy	(116)
Nivolumab, ipilimumab		IV/recurrent NSCLC	NCT02477826	1,739	396	29.3	0.2% myocarditis; 0.2% cardiac tamponade	(117)
Nivolumab, ipilimumab		IV NSCLC + brain met	NCT02696993	13	13	23	8% myocarditis (+ thoracic radiotherapy)	(118)
Nivolumab, relatlimab	PD-1, LAG-3	IB/II/IIIA NSCLC	NCT04205552	60	30	12	7% atrial fibrillation; 3% embolism	(21)

This table summarizes 19 prospective clinical trials (11,383 patients; follow-up 10.1–60 months), covering all lung cancer stages (stage IB–IV NSCLC, limited/extensive-stage SCLC) and eight ICI regimens, including PD-1/PD-L1 monotherapy and combination strategies. The incidence of cardiovascular irAEs ranges from 0.17% to 8%, with hypertension (1.3–6%) and vasculitis (0.17–3%) being the most common events. Myocarditis is relatively uncommon with ICIs alone (0.2–1.1%), and its incidence rises significantly to 8% with thoracic radiotherapy, which is associated with high mortality and requires close clinical attention. Dual ICI combination therapy is associated with a slightly higher risk of myocarditis compared with monotherapy. Differences in irAE incidence are also observed across ICI subclasses. Data on early-stage lung cancer remain limited. CTLA-4, cytotoxic T-lymphocyte-associated protein 4; ES-SCLC, extensive-stage small cell lung cancer; ICI, immune checkpoint inhibitor; irAEs, immune-related adverse events; LAG-3, lymphocyte activation gene 3; NSCLC, non-small cell lung cancer; PD-1, programmed death receptor 1; PD-L1, programmed death-ligand 1; SCLC, small cell lung cancer.

Table 3. Retrospective clinical studies on ICI-related cardiovascular toxicity in lung cancer treatment.

ICIs	Target	Lung cancer type/stage	Total patients, n	ICI-treated patients, n	Follow-up (months)	ICI-related cardiovascular irAEs (%)	References
Nivolumab, pembrolizumab	PD-1	Stage IIIB/IV NSCLC	60	60	NA	6.7% pericardial effusion	(119)
Pembrolizumab, nivolumab		Incident lung cancer	25,573	743	5.5	1.8% myocarditis or pericarditis; 5.8% cardiac arrhythmia; 2.5% heart failure; 3.4% cardiovascular death	(120)
Pembrolizumab, nivolumab		Advanced NSCLC	713	713	16.8	1.1% myocardial infarction; 0.4% heart failure; 1.4% stroke	(121)
PD-1/PD-L1 inhibitors†	PD-L1, PD-1	III–IV adenocarcinoma/squamous cell carcinoma	315	315	42	3.5% myocarditis	(122)
PD-1/PD-L1/CTLA-4 inhibitors‡	PD-1, PD-L1, CTLA-4	Primary lung cancer (stage unspecified)	252	135	6	0.7% myocarditis; 18.5% arrhythmia; 1.5% valvular heart disease; 14.8% cardiomyopathy; 5.9% pericardial disease; 5.9% thromboembolic events	(16)
PD-1/PD-L1/CTLA-4 inhibitors§		III–IV NSCLC and SCLC	196	196	NA	4.6% myocarditis; 2% pericardial disease; 1.5% NSTEMI; 3.6% new-onset SVT	(25)
PD-1/PD-L1/CTLA-4 inhibitors¶		Advanced lung cancer (stage unspecified)	5,255	5,255	72	0.06% myocarditis; 1.1% myocardial infarction; 3.9% heart failure; 2.5% atrial fibrillation; 1.3% conduction disorders; 3.5% stroke	(123)

Summary of seven retrospective real-world studies evaluating ICI-related cardiovascular toxicities in lung cancer (total 32,364 patients; follow-up 5.5–72 months). This study included patients with stage III–IV NSCLC and SCLC, treated with PD-1/PD-L1/CTLA-4 inhibitors as monotherapy, dual immunotherapy combinations, and ICI-based combination regimens. The incidence of cardiovascular and immune-related adverse events ranged from 0.06% to 18.5%, with arrhythmias, pericardial disease and heart failure being the most common; myocarditis is rare, but occurs more frequently with dual immunotherapy. Rare events such as valvular heart disease and non-ST-segment elevation myocardial infarction have also been observed. Real-world data can complement clinical trial results, providing a basis for toxicity monitoring, identification of high-risk populations, and long-term cardiovascular management. ^†, atezolizumab, durvalumab, sintilimab, camrelizumab, tislelizumab, toripalimab; ^‡, nivolumab, pembrolizumab, atezolizumab, durvalumab, ipilimumab; ^§, nivolumab, pembrolizumab, atezolizumab, ipilimumab; ^¶, ipilimumab, nivolumab, pembrolizumab, atezolizumab, avelumab, and durvalumab. CTLA-4, cytotoxic T-lymphocyte-associated protein 4; ICI, immune checkpoint inhibitor; irAEs, immune-related adverse events; NA, not applicable; NSCLC, non-small cell lung cancer; NSTEMI, non-ST-segment elevation myocardial infarction; PD-1, programmed death receptor 1; PD-L1, programmed death-ligand 1; SCLC, small cell lung cancer; SVT, supraventricular tachycardia.

In prospective clinical trials summarized in Table 2, ICI-related cardiovascular irAEs remain relatively infrequent in lung cancer but show substantial heterogeneity across treatment regimens. These data provide an evidence base for cardiovascular toxicity surveillance and regimen selection in ICI-treated lung cancer patients.

As summarized in Table 3, retrospective real-world studies reveal a higher and more heterogeneous incidence of ICI-related cardiovascular toxicities in lung cancer than that reported in clinical trials, particularly with combination regimens. These data underscore the importance of long-term monitoring and risk stratification in routine clinical practice.

Clinical trials provide standardized data, whereas retrospective studies better reflect long-term risks in the real world. Taken together, these findings confirm that combination therapy results in additive toxicity.

The data indicate that toxicity correlates with follow-up duration, with hypertension reaching 6% over 35.6 months of follow-up. It suggests that studies with relatively short follow-up periods may underestimate the cumulative risk, highlighting the importance of long-term monitoring. As reviewed above, although the incidence of cardiotoxicity associated with ICI treatment in lung cancer is relatively low, it predominantly involves severe adverse events, particularly in combination therapies.

Graded management and prevention strategies for cardiotoxicity

Risk stratification of ICI-related cardiac toxicity

Patient-specific clinical characteristics and ICI treatment regimens are key stratification factors for ICI-related cardiotoxicity, with distinct subgroups at heightened risk. Specifically, patients with pre-existing cardiovascular diseases including hypertension, coronary artery disease, heart failure, and myocardial infarction are at an increased risk of developing immune-related cardiac adverse events during ICI therapy (124-126). Additionally, patients with lung cancer who have undergone thoracic radiotherapy are more prone to ICI-related pericardial disease, attributed to the synergistic effects of radiotherapy and immunotherapy (124).

Notably, a retrospective study revealed that ICI-related cardiac adverse events (e.g., myocarditis, pericarditis, arrhythmias, myocardial infarction) are more common in males than in females (127). Furthermore, data from an international multicenter study demonstrated that the overall incidence of ICI-M is 1.14%, with significant variations by treatment regimen: 0.5% in patients receiving anti-PD-1 monotherapy, 2.4% in those on anti-PD-L1 monotherapy, 3.3% in anti-CTLA-4 monotherapy users, 2.4% in patients treated with anti-PD-1/anti-CTLA-4 combinations, and 1% in those receiving anti-CTLA-4/anti-PD-L1 combinations (61).

These findings underscore the need for individualized risk assessment prior to ICI initiation, as stratification by clinical characteristics and treatment regimens directly informs subsequent management decisions.

Graded management

The core management of ICI-M involves prompt ICI discontinuation, stratified immunosuppression, and enhanced cardiac monitoring [including serial troponin/N-terminal pro-B-type natriuretic peptide (NT-proBNP) measurements and ECG follow-up] (128).

To standardize the adverse reactions related to ICI therapy, such as cardiotoxicity, several organizations have issued relevant guidelines, including ASCO [2021] (129), the Society for Immunotherapy of Cancer Toxicity Management Working Group (SITC; 2021) (128), ESMO [2022] (130), the European Society of Cardiology (ESC; 2022) (131), and the National Comprehensive Cancer Network (NCCN; 2025) (132).

Table 4 presents a comparative analysis of these guidelines, highlighting their similarities and differences and identifying innovative aspects to guide clinical practice. The guideline grading references the Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0 (133), released by the United States National Cancer Institute on November 27, 2017.

Table 4. Grading, management and treatment of ICI-related cardiotoxicity: a guideline-based summary.

Guideline [year]	Diagnostic strategy	Classification & treatment	Restart timing of ICI
ASCO [2021, update 2024]	Biomarker-driven screening; imaging when indicated	G1–G4 grading. G1: hold ICI + close monitoring; ≥G2: discontinue ICI + corticosteroids (1–2 mg/kg/d); refractory: methylprednisolone (1,000 mg/d × 3–5 d) + immunosuppressants	Not specified; requires individualized assessment of cardiac function recovery
SITC [2021]	Early hospitalization; CMR and biopsy encouraged for confirmation	No formal grading. Confirmed: Methylprednisolone (1,000 mg/d × 3–5 d) → prednisone (1–2 mg/kg/d, taper over 4–6 w)	Consider only after complete symptom resolution and multidisciplinary confirmation
ESMO [2022]	Integrated clinical + biomarker + imaging approach	No specific grading. Confirmed: methylprednisolone (500–1,000 mg/d × 3 d) → prednisone (1 mg/kg/d) (troponin ↓ ≥50% & stable)	Usually discontinued permanently; rechallenge only after expert multidisciplinary review
ESC [2022]	Imaging-centered diagnostic workflow; EMB recommended in uncertain cases	No specific grading. Suspected: discontinue ICI + ECG/cTn monitoring (EMB if needed); confirmed: methylprednisolone (500–1,000 mg/d ≥3 d) + cTn monitoring (second-line immunosuppressants for non-response); complicated: ICU/MDT care; mechanical circulatory support if needed; permanent ICI discontinuation	Mostly permanent discontinuation; restart requires oncology/cardio-oncology consultation
NCCN [2025]	Surveillance-oriented; integrates biomarker monitoring with treatment decision	G1–G4 grading. G1: continue ICI; G2: discontinue ICI + methylprednisolone (1–4 mg/kg/d × 3–5 d); G3–4: methylprednisolone (500–1,000 mg/d × 3–5 d)	G2: may restart cautiously after baseline recovery; ≥ G3: permanent discontinuation

This table summarizes the grading and management recommendations extracted from selected international guidelines. Selection was based on publication recency, global relevance, and the inclusion of cardiotoxicity-specific recommendations. This compilation is not exhaustive, and further insights may be obtained from other regional or national guidelines. ASCO, American Society of Clinical Oncology; CMR, cardiovascular magnetic resonance; d, days; EMB, endomyocardial biopsy; ESC, European Society of Cardiology; ESMO, European Society for Medical Oncology; ICI, immune checkpoint inhibitor; ICU, intensive care unit; MDT, multidisciplinary team; NCCN, National Comprehensive Cancer Network; SITC, Society for Immunotherapy of Cancer; w, weeks.

In clinical practice, when applying various immune-toxicity management guidelines, flexible adjustments are necessary based on the specific conditions. The value of guidelines lies in providing a unified and scientific decision-making framework for clinical practice. However, given the complex underlying disease in patients with lung cancer, differences in immunotherapy regimens, and the diversity of toxicity types, it is essential to emphasize the individualized translation between “guidelines and patients”. For immune-related cardiotoxicity, in addition to assessing severity using grading systems such as CTCAE v5.0 (133) or ESC [2022] (131), a comprehensive assessment should be made by integrating dynamic monitoring indicators and multidisciplinary consultation.

Furthermore, there are certain differences among guidelines regarding grading criteria, initial doses of glucocorticoids, and recommendations on whether to restart ICI therapy. Clinicians should weigh the patient’s antitumor benefits against cardiovascular risks and formulate individualized strategies. Future guideline optimization should emphasize precise evaluation, early identification, and risk-prediction models for treatment resumption, thereby truly achieving patient-centered, evidence-based management. Once cardiotoxicity occurs, immunotherapy should generally not be restarted; however, a comprehensive assessment of the irAE risk based on the patient’s physical condition is necessary before considering restarting ICI treatment (134,135).

Preventive strategies

Given the limited regenerative capacity of the heart after injury, which is often irreversible, preventing cardiac toxicity induced by ICIs in lung cancer treatment is particularly crucial.

Before initiating ICI therapy, all patients should undergo baseline cardiovascular assessment, including ECG, echocardiography, troponin, and B-type natriuretic peptide (BNP) testing, to facilitate individualized surveillance strategies (136). In high-risk patients, particularly those with hypertension, coronary artery disease, or prior irAEs, closer monitoring with serial troponin measurements and optimization of cardiovascular medications should be considered (137). In addition, for patients suspected of ICI-related cardiac toxicity, prompt diagnostic evaluations should be performed for early diagnosis and treatment. Once ICI-induced cardiac toxicity is confirmed, ICIs should be discontinued immediately, and a comprehensive assessment should be performed to determine whether they should resume treatment (129).

Collectively, these measures can reduce the incidence of ICI-induced cardiac toxicity (Figure 3).

Figure 3.

Common clinical manifestations, diagnosis, and graded management of cardiac toxicity from ICI in lung cancer treatment. Common diagnostic methods include assessment of cardiovascular symptoms (such as chest tightness, shortness of breath), laboratory blood tests, ECG, ultrasound, CT, CMR, and myocardial biopsy. Treatment modalities include discontinuing ICIs, administering glucocorticoids, using IS, providing symptomatic treatment, and conducting MDT consultations. BNP, B-type natriuretic peptide; CMR, cardiovascular magnetic resonance; CRP, C-reactive protein; CT, computed tomography; ECG, electrocardiogram; ICIs, immune checkpoint inhibitors; IS, immunosuppressant; MDT, multidisciplinary team.

Biomarkers and predictive models for ICI-related cardiotoxicity

Early identification of ICI-M has long been a clinical challenge, attributed to its rapid progression and nonspecific manifestations, which often delay intervention. Compounding this issue, the high mortality of ICI-M is directly linked to inadequate early screening. Among ICI-M patients, 20.6% progress to complete heart block (CHB), and 86.5% of myocarditis-related deaths within six months are attributed to CHB (138). To address this dilemma, recent research has focused on two key strategies: identifying reliable early-warning biomarkers and developing precise risk-prediction models.

Cardiac-specific biomarkers, such as high-sensitivity cardiac troponin T (hs-cTnT) and NT-proBNP, which may increase 1–3 weeks after ICI administration even in asymptomatic patients (139,140), combined with systemic inflammatory markers, including neutrophil-to-lymphocyte ratio (NLR), C-reactive protein (CRP), and systemic immune-inflammation index (SII) (25,141-144) and dynamic ECG abnormalities [e.g., ST-T changes (30)], can effectively detect subclinical myocardial injury. Furthermore, integrating baseline cardiovascular comorbidities and oncologic treatment-related risk factors can construct validated nomograms for stratifying myocarditis severity and long-term survival outcomes (145), while combining these variables with thrombosis- or immune-related risk scores has yielded robust nomogram-based and multivariate predictive models for myocarditis and other irAEs (139,146-149). To synthesize recent advances (25,30,138,139,141-155), representative biomarkers and risk models for ICI-related cardiotoxicity are summarized below (Table 5).

Table 5. Emerging diagnostic biomarkers and predictive models for ICI-related cardiotoxicity.

Category	Biomarker/model	Clinical significance/diagnostic role	Key findings/evidence	References
Cardiac injury & function	ECG abnormalities, CHB, NT-proBNP	Early detection and severity assessment of ICI-M	ECG changes and elevated NT-proBNP associated with severe myocarditis and mortality	(30,138)
Peripheral inflammation & immune indexes	ANC, WBC, SII, SIRI, NLR, PLR, CRP	Reflect systemic inflammatory response and immune activation; early screening and severe risk prediction of cardiac irAEs in NSCLC	ANC, WBC, SII are valuable for diagnosing acute myocarditis; SII, SIRI, NLR, PLR and CRP correlate with adverse outcomes and cardiac irAE risk	(25,141-144)
Inflammatory cytokines and immune cell markers	IL-17A, Th17, Treg, GM-CSF, IL-6, TGF-β1	Mechanistic and diagnostic biomarkers	Proinflammatory immune imbalance associated with cardiac injury	(150,151)
Combined predictive models	NLR + PLT + LYM	Predict irAEs occurrence in NSCLC patients receiving ICIs	Higher LYM and lower NLR/PLT indicate increased irAE risk, including myocarditis	(152)
Composite severity markers	LAR, NHR, ΔLAR, ΔNLR	Severity grading and short-term prognosis	Combined dynamic markers predict MACEs and poor survival	(153)
Predictive nomograms/models	NT-proBNP + cTnI + LAR + SIRI; LDH + LAR + cTnI + AAR	Survival and mortality prediction	Multivariable models show high predictive accuracy	(154,155)
Integrated risk models	Baseline cardiovascular comorbidities + tumor treatment factors	Pre-ICI baseline cardiac risk stratification, predict myocarditis severity and long-term survival	Coronary artery disease predicts severe myocarditis and 90-day MACE; angiogenesis inhibitors increase severe myocarditis risk; thoracic radiotherapy indicates favorable prognosis.	(145)
AI-assisted diagnostic & multimodal strategies	AI-based ECG/CMR, time-series models; biomarkers + models + AI	Early diagnosis, treatment guidance, early warning and prevention	AI improves detection efficiency and outcome prediction; integrated approaches reduce fatal cardiac irAEs	(146-149)

This table summarizes emerging biomarkers and predictive models for early detection, severity assessment, and risk stratification of immune checkpoint inhibitor-related cardiotoxicity, particularly myocarditis. These include cardiac injury markers, systemic immune-inflammatory indices, cytokines, composite laboratory-based nomograms, and AI-assisted diagnostic tools. Together, they reflect a transition toward multimodal and precision-based strategies for early warning and clinical management of cardiac irAEs. AAR, albumin-to-alkaline phosphatase ratio; AI, artificial intelligence; ANC, absolute neutrophil count; CHB, complete heart block; CMR, cardiovascular magnetic resonance; CRP, C-reactive protein; ECG, electrocardiogram; GM-CSF, granulocyte-macrophage colony-stimulating factor; irAEs, immune-related adverse events; LAR, lactate dehydrogenase-to-albumin ratio; ICI-M, immune checkpoint inhibitor-related myocarditis; LDH, lactate dehydrogenase; LYM, lymphocyte; MACE, major adverse cardiovascular events; NHR, neutrophil-to-high-density lipoprotein cholesterol ratio; NLR, neutrophil-tolymphocyte ratio; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PLR, platelet-to-lymphocyte ratio; PLT, platelet; SII, systemic immune-inflammation index; SIRI, systemic inflammation response index; TGF-β1, transforming growth factor-β1; Treg, regulatory T cells; WBC, white blood cell count.

Collectively, current risk stratification is evolving from isolated biomarkers toward integrated and artificial intelligence (AI)-assisted multimodal prediction frameworks; however, external validation across large, prospective cohorts remains limited.

Advances in proteomics and genomics have enabled the identification of novel biomarkers that facilitate early diagnosis and monitoring of ICI-M. Concurrently, genomic and transcriptomic analyses have uncovered gene signatures associated with individual susceptibility and immune responsiveness to ICIs, offering new insights into the molecular basis of cardiotoxicity and supporting personalized therapeutic approaches (156-158).

Although significant progress has been made in this field, the specificity and sensitivity of novel biomarkers remain limited, and the lack of validation in large-scale cohorts means that clinical translation still faces considerable challenges. Recent research trends indicate that cardiac toxicity monitoring is shifting from passive detection towards active prediction. The integration of multidimensional biomarkers with AI-assisted ECG and cardiac magnetic resonance (CMR) imaging analysis offers a promising approach for optimizing diagnosis and implementing personalized monitoring (149,159). In summary, these advances have driven the development of precision cardiac oncology; by leveraging early screening and predictive models, it is hoped that safer and more personalized immunotherapy strategies can be established.

Discussion

This review proposes a lung cancer-specific susceptibility framework for ICI-related cardiotoxicity, moving beyond the perspective of pan-cancer research. It shows how the TME and immune inflammation contribute to cardiac injury, based on evidence from 26 studies including 43,747 patients across clinical trials and real-world cohorts. It also integrates emerging biomarkers and predictive models into a practical approach for early detection and risk stratification.

This study also has several limitations. Most of the included studies were retrospective or observational, which are prone to confounding factors and selection bias. Furthermore, definitions and grading criteria for cardiotoxicity varied across studies, making direct cross-comparisons between them difficult. In addition, prospective validation of biomarkers and artificial intelligence models remains insufficient, especially validation in specific populations of lung cancer patients. At the same time, evidence regarding the relevant mechanisms of action is largely confined to the preclinical stage or is based solely on studies with small sample sizes, lacking direct supporting clinical data.

Overall, cardiotoxicity induced by ICIs in patients with lung cancer is a multi-system interactive pathological process driven by tumor biological characteristics, host cardiovascular baseline susceptibility, and immune checkpoint-mediated immune reprogramming. Chronic inflammation, tissue hypoxia, previous thoracic radiotherapy, and persistent immune activation jointly create a susceptible microenvironment, which further induces endothelial dysfunction and triggers abnormal immune responses. On this basis, ICI treatment can further amplify the systemic immune response. In some cases, it may exacerbate myocardial injury through tumor-cardiac antigen cross-reactivity, ultimately leading to clinical cardiotoxicity.

This framework supports a shift from single-mechanism explanations to an integrated cardio-oncology model, where tumor- and patient-related factors jointly determine cardiac risk and clinical phenotype. It provides a basis for more precise risk stratification, biomarker-guided monitoring, and individualized management strategies.

Future work should validate this lung cancer-specific framework in large prospective cohorts, standardize diagnostic criteria, and further evaluate biomarkers and AI-assisted models. Development of targeted cardioprotective and immunomodulatory strategies for high-risk patients remains an important goal.

Conclusions

ICIs have revolutionized the treatment landscape for advanced lung cancer; however, immune-related cardiotoxicity has emerged as a clinically significant and potentially life-threatening complication, significantly impacting treatment safety and long-term prognosis. Myocarditis is the most serious form of this condition, highlighting the need for early identification and prevention.

Lung cancer patients are at high risk of cardiotoxicity due to pre-existing cardiovascular disease, smoking-related inflammation, prior thoracic radiotherapy, and TME remodeling, which together contribute to endothelial dysfunction and immune activation. Combination immunotherapy and radiotherapy may further increase this risk. Mechanistically, ICI-related cardiotoxicity reflects immune-vascular-myocardial interactions rather than isolated organ injury.

In clinical practice, pre-treatment cardiovascular risk assessment and baseline cardiac evaluation are essential. Ongoing surveillance can integrate cardiac biomarkers, including NT-proBNP and hs-cTnT, systemic inflammatory markers such as NLR, CRP, and SII, and functional assessment using ECG and CMR, to enable early detection of subclinical cardiotoxicity. Multidisciplinary cardio-oncology collaboration and guideline-based management strategies are needed to support timely risk stratification and treatment adjustment.

Future research should focus on establishing lung cancer-specific prospective registries with standardized assessment protocols, validating biomarkers and AI-assisted predictive models in large cohorts, and developing evidence-based cardioprotective and immunomodulatory strategies for high-risk patients. These efforts are critical for building a standardized, lung cancer-specific cardio-oncology care pathway that integrates prevention, monitoring, and intervention.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.