Precision Cardio-Oncology: A Multimodal Approach to Cancer Therapy-Related Cardiovascular Toxicity Prediction and Assessment.
Authors
Affiliations (12)
Affiliations (12)
- Cardio-Oncology Service, University College London Hospitals NHS Foundation Trust, London NW1 2PG, UK.
- Hatter Cardiovascular Institute, University College London, London WC1E 6HX, UK.
- Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London E1 2AD, UK.
- Faculty of Medical Sciences, University College London, London WC1E 6BT, UK.
- Velindre Cancer Centre, Velindre University NHS Trust, Cardiff CF14 2TL, UK.
- Global Business School for Health, Faculty of Population Health Sciences, University College London, London E20 2AE, UK.
- Department of Cardiology, London North West University Healthcare NHS Trust, London HA1 3UJ, UK.
- National Heart and Lung Institute, Faculty of Medicine, Imperial College London, London SW3 6LY, UK.
- Oncology and Cardio-Oncology Services, Nottingham University Hospitals NHS Trust, Nottingham NG7 2UH, UK.
- Institute of Health Informatics, Faculty of Population Health Sciences, University College London, London NW1 2DA, UK.
- Cardio-Oncology Service, Barts Heart Centre, Barts Health NHS Trust, London EC1A 7BE, UK.
- William Harvey Research Institute, NIHR Barts Biomedical Research Centre, Faculty of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK.
Abstract
Cancer therapy-related cardiovascular toxicity (CTR-CVT) is associated with morbidity, mortality, and interruptions to cancer treatment. Current cardio-oncology clinical assessment tools generally rely on a limited range of clinical data; however, in the era of precision medicine, it may be possible to leverage multimodal data, including imaging and multi-omic inputs, to personalise CTR-CVT prediction and assessment. In this review, we examine the genotypic and phenotypic predictors of CTR-CVT and the evidence supporting their incorporation into multimodal risk models. Most research focuses on cancer therapy-related cardiac dysfunction (CTRCD). We identify a strong evidence base for multimodal imaging-guided prediction, particularly cardiovascular magnetic resonance (CMR) and transthoracic echocardiography (TTE)-derived global longitudinal strain measurements. Emerging evidence also supports the potential value of novel imaging parameters, including computed tomography (CT) and TTE-derived radiomic features. From the genomic and proteomic perspective, truncating titin variants and microRNA-126 may be informative predictors of cancer therapy-related cardiac dysfunction (CTRCD). At the same time, the literature also highlights potential epigenomic markers of CTR-CVT. Finally, we consider how these diverse data can be integrated into multimodal risk models and how digital innovation and advanced machine learning methods may further improve model predictive performance. However, a large implementation gap remains, with challenges such as data privacy and a lack of external validation studies preventing routine clinical implementation.