Radiation-induced cardiotoxicity in lung cancer: Current landscape and future directions

Abstract

Thoracic radiotherapy is crucial for lung cancer treatment but poses a high risk of radiation-induced heart disease (RIHD), a major cause of non-cancer health issues and death in survivors. This risk is heightened by cardiovascular risk factors, the use of combined therapies, and better long-term survival rates. This review explores RIHD in lung cancer, emphasizing the significant epidemiological impact, especially for left-sided tumors, and identifies dose-volume predictors for cardiac areas like the left anterior descending coronary artery and left ventricle. It discusses the pathophysiology of chronic inflammation, microvascular damage, and fibrosis, and examines the use of advanced predictive modeling, including radiomics and machine learning, for personalized risk assessment. These methodologies, initially developed for radiation pneumonitis, offer a reliable framework for predicting thoracic toxicities like RIHD due to shared biological processes, including inflammation and fibrosis. The interactions between radiotherapy and new systemic agents are examined, highlighting the importance of careful cardio-oncology management. Current mitigation strategies include heart-sparing planning, pharmacological interventions, and early detection methods like global longitudinal strain. We suggest future directions such as routine substructure-based planning, validation of multi-omics predictive models, and establishing protocol-driven cardio-oncology pathways to optimize tumor control and maintain cardiovascular health in lung cancer patients.

Key Words: Radiation-induced heart disease; Lung cancer; Cardiotoxicity; Dose-volume parameters; Immunotherapy; Cardio-oncology

Core Tip: This mini-review examines the evolving comprehension of radiation-induced cardiotoxicity in lung cancer, emphasizing the transition from whole-heart to cardiac substructure-focused dose evaluation and risk prediction. The primary focus is on utilizing advanced predictive modeling techniques, such as radiomics and machine learning, to achieve personalized risk stratification. Importantly, the methodological paradigm developed for predicting radiation pneumonitis is highlighted as a crucial surrogate framework for creating analogous models for cardiotoxicity, given the similarities in injury mechanisms and data-driven modeling approaches. Additionally, the review underscores the essential need for integrated cardio-oncology management strategies in the context of combination therapies involving immunotherapy and targeted agents, with the goal of maintaining long-term cardiovascular health without compromising oncologic outcomes.

INTRODUCTION

Thoracic radiation remains a cornerstone in the curative and palliative treatment of lung cancer, the largest cause of cancer-related death globally. Survival rates for people with non-small cell lung cancer (NSCLC) have progressively increased because of advances in treatment techniques, such as immunotherapy and targeted treatments. However, this paradigm shift has highlighted the long-term implications of cancer therapy, with radiation-induced cardiotoxicity standing out as a substantial and potentially life-threatening problem. Cardiovascular illness is quite common in the lung cancer community, impacting 30%-50% of patients and accounting for approximately 30% of fatalities, a burden that reflects both shared risk factors like smoking and the direct cardiotoxic potential of thoracic irradiation[1,2].

The historical perception of the heart as a radioresistant organ has been fundamentally challenged by a growing body of evidence that links radiation dose to cardiac structures with adverse outcomes. Recent studies have demonstrated that patients receiving radiotherapy for thoracic malignancies are at significantly increased risk of cardiovascular-related death compared with the general population[3]. This risk is not uniform; it exhibits a distinct spatial dependence, being significantly higher for tumors in the left lower lung compared with the upper lung or right chest. The cardiotoxic effects of radiation are now understood to manifest across a spectrum, including pericardial disease, cardiomyopathy, coronary artery disease, valvular dysfunction, and conduction abnormalities, collectively termed radiation-induced heart disease (RIHD).

The clinical imperative to understand, predict, and mitigate RIHD is driven by the dual goals of optimizing cancer control and preserving long-term cardiovascular health. This challenge is compounded by the evolving treatment landscape, where radiotherapy is increasingly combined with systemic agents like immune checkpoint inhibitors (ICIs) and tyrosine kinase inhibitors, which may have independent or synergistic cardiovascular effects[4]. Furthermore, the shift towards hypofractionated regimens and advanced techniques like intensity-modulated radiotherapy and proton therapy necessitates a continuous re-evaluation of dose-volume relationships and toxicity risks[5,6]. This review synthesizes the current epidemiological and clinical burden of radiation-induced cardiotoxicity, elucidates its underlying pathophysiology, examines cardiac substructure-specific dose predictors, advanced predictive modeling, interactions with novel systemic therapies, and contemporary mitigation strategies, culminating in a perspective on future directions to improve the therapeutic ratio for lung cancer patients.

THE EPIDEMIOLOGICAL AND CLINICAL BURDEN OF RADIATION-INDUCED CARDIOTOXICITY IN LUNG CANCER

Radiation-induced cardiotoxicity in lung cancer has a significant epidemiological footprint and is increasingly acknowledged as a factor influencing long-term survival and quality of life. Large cohort analyses show that thoracic radiation is associated with an increased risk of cardiovascular morbidity and mortality[3]. In locally advanced NSCLC, the 2-year cumulative incidence of serious cardiac events, including heart failure and acute coronary syndromes, is high, and severe arrhythmias are common, often occurring years after therapy[7,8].

The burden of cardiotoxicity is not distributed evenly across patient subgroups or tumor locations. A critical finding is the lateralization of risk. Patients receiving radiotherapy for tumors in the left chest, particularly the left lower lung, exhibit a significantly higher risk of RIHD compared with those with right-sided tumors, with hazard ratios indicating an 11% higher risk for left lower lung cancers compared with left upper lung cancers[3]. This anatomical vulnerability is likely due to the proximity of left-sided tumors to critical cardiac substructures like the left anterior descending coronary artery and the left ventricle. The clinical impact of these cardiac events is profound. Patients who develop severe radiation pneumonitis (RP) and experience unsuccessful steroid tapering due to rebound of symptoms have significantly shorter progression-free and overall survival compared with those with successful tapering, highlighting how thoracic toxicities can intersect and impact oncologic outcomes[9]. Moreover, the prevalence of pre-existing cardiovascular comorbidities is high in lung cancer patients, a legacy of shared risk factors like smoking, which further amplifies their vulnerability to treatment-related cardiac injury[10]. This complex backdrop requires a detailed understanding of the biological mechanisms that underpin radiation damage to the heart.

MECHANISMS AND PATHOPHYSIOLOGY OF RIHD

The pathophysiology of RIHD is multifaceted, involving direct cellular damage and complex cascades of inflammation and fibrosis that unfold over time. Ionizing radiation causes direct DNA damage in cardiac cells, including cardiomyocytes, endothelial cells, and fibroblasts, triggering cellular senescence, apoptosis, and a persistent pro-inflammatory state. In the heart, this direct injury has unique consequences. Cardiomyocytes, being largely post-mitotic, are particularly susceptible to radiation-induced mitochondrial dysfunction and oxidative stress, which can lead to irreversible cell death and compromised contractile function. Cardiac microvascular endothelial cells are also key targets; their damage disrupts the coronary microcirculation, leading to perfusion defects, chronic ischemia, and a pro-fibrotic microenvironment. This initial injury starts a chronic process characterized by microvascular damage, perfusion defects, and oxidative stress, which ultimately leads to myocardial fibrosis and remodeling. Preclinical models are instrumental in dissecting these pathways. In vivo studies using mice with pre-existing interstitial lung disease that underwent partial thoracic irradiation demonstrated severe, bilateral diffuse alveolar damage with overlapping exudative, proliferative, and fibrosing patterns, accompanied by hyperactivation of inflammatory responses and infiltration of macrophages and CD4⁺ lymphocytes[11]. This underscores the role of pre-existing inflammation in exacerbating radiation injury.

A key mediator in this process is the cytokine milieu. Transforming growth factor-beta (TGF-β) is a central driver of fibrosis, promoting the differentiation of fibroblasts into collagen-secreting myofibroblasts. Other cytokines and chemokines, such as tumor necrosis factor-alpha (TNF-α) and monocyte chemoattractant protein-1, perpetuate inflammation and recruit immune cells to the injured tissue. Recent research has begun to delineate a more integrated causal network among these mechanisms. The initial radiation-induced DNA damage and reactive oxygen species generation in endothelial cells not only cause apoptosis, but also upregulate adhesion molecules such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1, facilitating leukocyte infiltration[12]. This sustained inflammatory infiltrate, rich in macrophages and T-lymphocytes, becomes a major source of pro-fibrotic cytokines like TGF-β and pro-inflammatory mediators such as interleukin-1β and TNF-α[13-15]. Notably, TGF-β acts as a critical signaling hub: It is both induced by and amplifies the inflammatory response, while simultaneously driving the transition of resident cardiac fibroblasts into activated myofibroblasts. This process creates a vicious cycle where inflammation promotes fibrosis, and fibrotic tissue in turn perpetuates a low-grade inflammatory milieu due to impaired vascularization and hypoxia. Within the cardiac context, this sustained inflammatory response and fibrotic cascade have organ-specific manifestations. Radiation-induced cardiac fibrosis tends to be diffuse and interstitial, leading to myocardial stiffening, diastolic dysfunction, and eventual heart failure[16]. Furthermore, radiation accelerates atherosclerosis in the coronary arteries, but with a distinct pattern often involving ostial and proximal segments, potentially related to direct endothelial injury and vasa vasorum dysfunction[15]. The conduction system is also vulnerable; fibrosis of the sinoatrial node, atrioventricular node, or the His-Purkinje network underlies various radiation-induced arrhythmias. Molecular studies have identified specific signaling pathways involved in malignant transformation following radiation exposure. Research on radiation-induced lung cancer models found that malignant transformation is accompanied by chromatin switching and the generation of tumor-specific chromatin loops mediated by TP63, which underlies MYC oncogene activation[17]. While this study focused on carcinogenesis, it illustrates the profound and lasting genomic and epigenetic alterations induced by radiation, which may have parallels in the pathogenesis of cardiac fibrosis and dysfunction, such as the persistent activation of pro-fibrotic gene programs in cardiac fibroblasts.

Emerging research implicates microvascular damage as a fundamental upstream event that connects acute injury to late fibrosis. Endothelial cell dysfunction reduces nitric oxide bioavailability, increases vascular permeability, and activates the coagulation cascade, resulting in a pro-thrombotic state and capillary rarefaction. Chronic ischemia and hypoxia significantly increase TGF-β expression and fibroblast activation. Furthermore, recent preclinical studies have highlighted the role of specific innate immune pathways, such as the NLRP3 inflammasome and the cyclic GMP-AMP synthase-stimulator of interferon genes pathway, in sensing radiation-induced cellular damage (mitochondrial DNA release) and amplifying the sterile inflammatory response that drives this pathological cascade[18].

The pericardium frequently acts as an early responder to cardiac radiation injury. Computed tomography (CT)-based studies investigating pericardial composition changes demonstrated that radiation induces dose-dependent changes in pericardial tissue composition that can be seen on standard-of-care CT scans approximately 6 months after radiation therapy. These alterations, which included increases in fibrous tissue and effusion components, were strongly related with a shorter time to a post-RT cardiovascular disease diagnosis, suggesting that they could serve as early imaging indicators for late cardiotoxicity[19]. This growing understanding of cardiac-specific cellular and molecular mechanisms, particularly the interconnected network of endothelial dysfunction, chronic inflammation, and fibrotic remodeling, guides the search for specific dose-volume relationships at the cardiac substructure level, which is critical for improving treatment planning and risk prediction.

The interconnected network of endothelial dysfunction, chronic inflammation, and fibrotic remodeling, as described above, is summarized in Figure 1. This proposed mechanistic cascade highlights the dynamic and self-perpetuating nature of RIHD pathogenesis, moving from initial radiation insult to chronic cardiac compromise.

Figure 1 Schematic diagram of the pathophysiological cascade of radiation-induced heart disease. The illustration depicts the dynamic progression from acute injury to chronic cardiac dysfunction following thoracic radiotherapy, encompassing four interconnected stages. A: Initial injury and endothelial dysfunction; B: Chronic inflammation and immune activation; C: Fibrotic remodeling; D: Clinical phenotypes. VCAM: Vascular cell adhesion molecule; ICAM: Intercellular adhesion molecule; ROS: Reactive oxygen species; TNF-α: Tumor necrosis factor-alpha; TGF-β: Transforming growth factor-beta; IL-1β: Interleukin-1β; MCP-1: Monocyte chemoattractant protein-1; cGAS: Cyclic GMP-AMP synthase; STING: Stimulator of interferon genes.

CARDIAC SUBSTRUCTURES AND DOSE-VOLUME PREDICTORS OF SPECIFIC TOXICITIES

The paradigm for cardiac dose constraints has evolved from evaluating the whole heart to focusing on specific substructures, as different regions of the heart exhibit varying radiosensitivity and are associated with distinct clinical toxicities. This granular approach is critical for accurate risk assessment and planning optimization. A pivotal study demonstrated that dose-volume histogram parameters of cardiac substructures had superior predictive power and robustness over whole-heart variables for predicting early elevations in high-sensitivity cardiac troponin T, a biomarker for cardiac injury. The volume receiving at least 20 Gy dose (V20 Gy) in the left anterior descending coronary artery was identified as the most dominant predictor[20]. This finding aligns with the clinical observation that coronary artery disease is a major manifestation of RIHD.

Particular toxicities are associated with the irradiation of specific substructures. Doses to the left ventricle correlate with the risk of heart failure, while exposure of the left anterior descending artery correlates with coronary events, and doses to various coronary vessels and pulmonary veins correlate with specific arrhythmias[7,8]. These correlations indicate pathophysiologically distinct processes depending on the irradiation substructure.

Substructure-specific interactions have direct ramifications for survival. A comprehensive, multi-institutional study of 1587 stage III NSCLC patients revealed that although cardiovascular dose parameters exhibited significant correlations with overall survival, their additional contribution to predictive models beyond clinical and lung dosimetric parameters was minimal with modest effect sizes. While protecting the heart is important, it must be balanced with other clinical factors[21]. The implementation of these dosage predictors in clinical practice is facilitated by sophisticated treatment planning methods that effectively preserve these essential areas, a fundamental aspect of contemporary mitigation tactics.

ADVANCED PREDICTIVE MODELING AND RISK STRATIFICATION

The integration of advanced computational methods, particularly radiomics, dosiomics, and machine learning, represents a promising but still largely investigational frontier in personalized risk prediction for radiation-induced toxicities, moving beyond traditional dose-volume parameters. It is critical to distinguish between established clinical dosimetric parameters and these emerging, data-driven approaches. Most current evidence supporting radiomics and machine learning models is derived from retrospective, single-center studies that often lack rigorous external validation and demonstration of clinical utility in prospective settings. While these approaches can extract high-dimensional quantitative features from medical images and dose distributions to build predictive models with intriguing accuracy, their translational readiness is currently limited. The following discussion highlights both the potential and methodological challenges of these tools.

While significant foundational research in this domain has concentrated on RP, the established modeling frameworks, validation methodologies, and understandings of feature significance are directly pertinent and applicable to the prediction of radiation-induced cardiotoxicity because they share common pathophysiological pathways involving radiation-induced inflammation, vascular injury, and fibrotic remodeling. Therefore, the high predictive performance demonstrated for pneumonitis models provides a strong methodological precedent and proof-of-concept for building analogous models for cardiac risk. Meta-analyses show that these models work very well for RP, and integrated radiomics-clinical models have a high level of predictive accuracy[22]. The effectiveness of these multi-feature, data-integration strategies for RP highlights their prospective applicability in cardiotoxicity prediction, where the integration of cardiac-specific imaging features, dose distributions to substructures, and clinical covariates may produce comparably robust models. It is important to note that these impressive numbers come from research settings, and they will not be widely used in clinical settings until they are standardized and tested. RP largely pioneered these methodologies, providing a conceptual and computational framework for cardiotoxicity prediction.

Combining different forms of data increases prediction potential. For example, multi-omics techniques can predict numerous toxicities at once, and models that combine radiomic, dosiomic, deep learning, and clinical characteristics have demonstrated excellent discriminatory performance in validation cohorts[23,24].

Prediction is further improved by functional imaging characteristics. When compared with typical anatomical dose metrics, research using 4D-CT to create ventilation and perfusion maps has demonstrated that dose-function metrics, such as the mean dose to well-ventilated lung, have a greater prognostic potential for pneumonitis[25,26]. Additionally, artificial intelligence can automatically detect risk variables that are significantly linked to severe RP, such as interstitial lung abnormalities on CT[27,28]. The ultimate objective is to convert these complex models into clinical decision support tools that facilitate the identification of high-risk patients prior to treatment and direct the modification of personalized plans, therefore averting major consequences.

While the reported performance metrics of radiomic and machine learning models are promising, a critical synthesis reveals significant heterogeneity and methodological challenges that preclude immediate clinical adoption. First, there is a notable lack of standardization in feature extraction, model validation, and performance reporting across studies, leading to difficulties in comparing results and assessing generalizability[22,29]. Many models are developed on single-institution datasets with limited sample sizes, and their performance often degrades upon external validation. Second, the biological interpretability of many high-dimensional radiomic features remains unclear, raising concerns about model robustness and clinical plausibility. Moreover, the majority of studies concentrate exclusively on predictive accuracy, overlooking essential factors for clinical application, including model calibration, decision-curve analysis for net benefit, and incorporation into the radiotherapy workflow. Therefore, while the shift toward data-driven prediction is clear, current evidence must be regarded as preliminary. Future research must prioritize large-scale, prospective multi-institutional efforts with standard imaging and contouring protocols to develop truly generalizable models and explicitly address the trade-offs between model complexity and clinical utility.

EVOLVING TREATMENT LANDSCAPES: INTERACTIONS WITH IMMUNOTHERAPY AND TARGETED THERAPIES

The integration of radiotherapy with novel systemic agents, particularly ICIs and targeted therapies, has improved outcomes but introduced new complexities in toxicity management[30], including potential interactions that affect cardiotoxicity risk. Understanding these interactions requires a dual perspective: First, examining direct cardiovascular toxicities of these agents; and second, considering how systemic and pulmonary toxicities may indirectly influence cardiovascular risk or share underlying mechanistic pathways with cardiac injury. While ICIs have become standard as consolidative therapy after chemoradiation for unresectable stage III NSCLC, they can induce inflammatory toxicities, including myocarditis, pericarditis, and vasculitis, which are of direct cardio-oncologic relevance. Furthermore, ICI-mediated chronic inflammation is hypothesized to accelerate atherosclerosis, a particular concern in multimorbid patients[1].

While early evidence came from retrospective analyses and small single-arm trials, recent prospective randomized controlled trials (RCTs) provide more robust insights into the safety and efficacy of combining radiotherapy with immunotherapy. The landmark PACIFIC trial, which established consolidative durvalumab after chemoradiotherapy as the standard of care for stage III NSCLC, prospectively demonstrated a manageable safety profile with no new or unexpected cardiac safety signals reported in the durvalumab arm compared with placebo, despite the inclusion of patients who had received thoracic radiotherapy[31]. Subsequent analyses from real-world cohorts and expanded access programs have largely corroborated these findings, suggesting that the cardiovascular risk profile of sequential chemoradiation and ICI consolidation may be primarily driven by the known toxicity of each modality rather than unexpected synergies[32,33].

Radiation and immunotherapy can work together to enhance anticancer immune responses while aggravating normal tissue inflammation. Preclinical models reveal overlapping pathways; for example, radiation and immunotherapy may have a synergistic effect on lipid metabolism dysregulation and oxidative stress in the tumor microenvironment, potentially affecting both pulmonary and cardiovascular damage[4]. Clinical research focused on combination modality toxicity provides vital information. The thorough study of RP in this setting provides an important model for understanding severe, immune-mediated normal tissue harm. The lessons obtained in dose constraint adjustment, risk prediction, and steroid-refractory inflammation management are conceptually relevant to cardio-oncology. A phase 2 trial of nivolumab with or without ipilimumab after concurrent chemoradiation found that lung V20 Gy > 23% was associated with a significantly higher rate of grade 2 or greater pneumonitis. This suggests that traditional dose constraints may need to be tightened in immunotherapy[34]. This principle that combination therapy may lower the toxicity threshold is relevant for cardiac tissues; however, direct cardiac dosage limitations in this scenario have yet to be determined.

Importantly, although they are scarcer, direct cardiovascular data from combination therapy are starting to appear. Pneumonitis was linked to a lower progression-free survival but not overall survival in another retrospective analysis of NSCLC patients treated with thoracic irradiation and ICIs[35]. Additionally, combination radiation and ICI pneumonitis did not always result in worse outcomes. The need for thorough toxicity care that includes cardiac surveillance is highlighted by the fact that severe inflammatory toxicity might affect oncologic outcomes, even though cardiotoxicity was not the study’s primary objective. There are currently no prospective trials explicitly intended to assess cardiac endpoints in this cohort. However, this vacuum is starting to be filled by prospective observational studies and secondary analyses of larger RCTs. The atherosclerosis acceleration theory has been supported in part by prospective cohort studies that track patients with multimorbidity receiving ICIs and report accelerated coronary calcium score progression.

For targeted therapies, such as third-generation EGFR tyrosine kinase inhibitors, combination treatment with radiotherapy significantly increases the incidence of RP[36]. More directly pertinent to the heart, osimertinib carries an independent risk of cardiotoxicity (such as QT prolongation and heart failure), and factors such as smoking history, hyperlipidemia, and concurrent left chest radiotherapy independently increase this risk[37]. The evidence base for the cardiotoxicity of radiotherapy combined with TKIs consists primarily of retrospective analyses and pharmacovigilance data. Large-scale prospective trials evaluating this specific combination are scarce, highlighting a significant evidence gap. Therefore, current management recommendations are extrapolated from the safety profiles of each individual treatment and expert consensus, emphasizing the critical need for prospective studies with protocolized cardiac monitoring in patients receiving these combinations.

The current understanding of interactions between radiotherapy and novel systemic agents is evolving and marked by areas of both promise and significant uncertainty. The potential for synergistic antitumor effects is counterbalanced by the risk of synergistic normal tissue toxicity, particularly pneumonitis and cardiotoxicity. However, the evidence base is characterized by retrospective analyses, small prospective trials, and a predominance of pneumonitis-focused endpoints, with cardiotoxicity often under-reported or assessed as a secondary outcome[4,35]. This leads to conflicting interpretations; some data suggests that traditional lung dose constraints may need tightening in the immunotherapy era[34], while other analyses find no clear survival detriment from combined-modality pneumonitis[35]. A major gap is the lack of prospective studies specifically designed to delineate the causal relationship and time course of cardiovascular events following combination therapy. It remains unclear whether these agents independently accelerate atherosclerosis, potentiate radiation-induced vascular injury, or simply unmask pre-existing subclinical disease. Consequently, current cardio-oncology management recommendations are largely extrapolated from evidence in other cancer types or based on expert opinion, highlighting an urgent need for dedicated prospective studies with rigorous cardiac phenotyping in lung cancer cohorts receiving modern combined modality treatments.

MITIGATION STRATEGIES: FROM TREATMENT PLANNING TO CARDIOPROTECTIVE INTERVENTIONS

Mitigating radiation-induced cardiotoxicity requires a multifaceted approach that includes sophisticated treatment planning, technical advances, pharmaceutical therapies, and constant monitoring. The core of mitigation is sophisticated treatment planning, which actively reduces doses to cardiac tissues. Techniques like heart-sparing volumetric modulated arc therapy and cardiac avoidance zones considerably reduce doses to essential structures while maintaining goal coverage[38-40].

Advanced methods such as proton beam therapy have inherent physical benefits for cardiac sparing. Studies indicate that intensity-modulated proton treatment is safe and may enhance survival by reducing cardiac dosage, particularly for cancers near the heart[41,42]. Online adaptive radiotherapy, particularly on systems like MR-Linacs, enables daily plan updates to account for anatomical changes, further enhancing cardiac sparing; however, this advantage must be balanced against the time required for adaptation, which lowers patient throughput[43].

Pharmacological cardioprotection is an emerging area. While traditional management of RP relies on corticosteroids, novel agents are being explored. The antifibrotic drug pirfenidone showed preliminary efficacy and tolerability in reducing radiation-induced lung injury in a retrospective pilot study[44]. Investigations into agents like pirfenidone for pulmonary fibrosis offer a parallel research pathway for identifying cardioprotective drugs, as they target the fibrotic process, a central mechanism in both RP and RIHD. In a randomized trial, the multi-tyrosine kinase inhibitor nintedanib was tested alongside steroids for RP treatment, with associated biomarker changes identified[45]. Other interventions, such as the herbal combination HL301, have been tested for safety but did not effectively reduce RP in a phase 2a trial[46]. For established cardiac dysfunction, echocardiographic global longitudinal strain has emerged as a sensitive, non-invasive tool for early detection of radiation-induced cardiac dysfunction, potentially enabling timely cardioprotective therapy[47]. These strategies collectively aim to preserve cardiovascular health without compromising oncologic efficacy.

CONCLUSION

Radiation-induced cardiotoxicity has rightfully ascended to a position of critical importance in the multidisciplinary care of lung cancer patients. The convergence of improved survival, an aging population with pre-existing cardiovascular risk, and the integration of potentially cardiotoxic systemic therapies has created a need to prioritize cardiac health alongside tumor control. The evidence is unequivocal: Thoracic radiotherapy significantly increases the risk of cardiovascular morbidity and mortality, with risks modulated by dose, volume, specific cardiac substructures irradiated, and patient-specific factors[48]. However, it is critical to interpret this evidence within its limitations. Key challenges include the heterogeneity of cardiac endpoint definitions across studies, the often incomplete or short-term nature of cardiac follow-up in oncology trials, and the competing risks of cancer progression and non-cardiac death in advanced lung cancer populations, which can obscure the full incidence and impact of RIHD.

The evolution from whole-heart dose metrics to substructure-specific parameters represents a significant conceptual advance, enabling more precise risk prediction and paving the way for tailored treatment planning. The future of managing this complex toxicity lies in the seamless integration of advanced technologies, predictive analytics, and collaborative clinical care. Prospective integration of cardiac substructure contouring and dose constraints into routine treatment planning for lung cancer is an important next step, which requires the development and widespread adoption of standardized contouring protocols, such as atlases from groups like the Radiation Therapy Oncology Group or the European Society for Radiotherapy and Oncology, to ensure consistency across clinical trials and practice. This effort should be guided by consensus guidelines derived from large-scale clinical data and prospective trials specifically designed to validate cardiac dose constraints, such as those embedded within modern cardio-oncology study frameworks.

The emerging sciences of radiomics and machine learning have enormous potential for developing personalized risk profiles. However, as previously noted, these are primarily investigative tools. Their path to clinical effect requires validation in multi-institutional studies, proof of superior results to existing models, and seamless integration into clinical workflow systems. Crucially, any prediction model must consider the competing mortality risks in this patient population. Furthermore, the differential impact of innovative fractionation schemes and particle therapy on cardiac substructures and late effects warrants additional investigation through long-term follow-up studies of clinical trials and specialized prospective research comparing modalities.

Perhaps the most crucial frontier is the establishment of proactive, protocol-driven cardio-oncology pathways. This includes baseline cardiovascular risk stratification, routine monitoring with sensitive biomarkers and imaging, and the investigation of targeted cardioprotective agents in randomized settings. Ongoing and future clinical trials evaluating cardioprotective strategies in the radiotherapy context are essential to translate mechanistic insights into clinical practice. The ultimate goal is to ensure that victories over lung cancer are not diminished by preventable cardiovascular sequelae, thereby securing both longevity and quality of life for cancer survivors. Achieving this goal requires a balanced view that acknowledges the robust evidence for risk, the promising but preliminary nature of many advanced mitigation strategies, and the persistent gaps in long-term outcome data.