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.

Keywords: 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.