Background: Treatment-induced cardiotoxicity is a significant concern in breast cancer patients undergoing multimodal therapy, surpassing cancer-related mortality as survival rates improve. Identifying patients at risk of cardiotoxicity early on is crucial for initiating preventive and management strategies to mitigate these effects. However, there is limited data on the incidence of cardiotoxicity.
Aims: The primary objectives of this study are to utilize the INCEPTION dataset to determine the prevalence and patterns of cancer therapy-related cardiac dysfunction (CTRCD) in Irish breast cancer patients receiving treatment. We aim to identify specific risk factors associated with both patients and treatments. Furthermore, we intend to integrate clinical and cardiac imaging data and employ Machine Learning (ML) techniques to develop a risk profiling model. This model will be instrumental in guiding management strategies.
Plan of Investigation: We will meticulously examine the INCEPTION clinical dataset to define the prevalence, patterns, and predictors of both oncologic and cardiovascular outcomes. This dataset was collected from breast cancer patients treated over a span of 12 years (2005-2017). We will achieve this by utilizing demographic, clinical, and treatment-specific data. Additionally, we will integrate cardiac imaging data with the INCEPTION clinical dataset to evaluate how echocardiography features can enhance our ability to predict CTRCD. Lastly, we will leverage the integrated dataset to create an ML-based risk model, which will be made accessible through web-based software systems. This model will stratify patients based on their risk of developing CTRCD, thereby informing decisions regarding surveillance, prevention, and treatment strategies.
Potential for Addressing Practice-Relevant Questions: By incorporating cardiac risk assessment into cancer treatment planning, we can proactively identify high-risk patients. This approach allows us to tailor treatment and surveillance strategies, ultimately reducing the likelihood of significant adverse cardiac events.