An Innovative CFD-based Dosimetry and Pre-treatment Planning Platform to Support Personalized Transarterial Therapies for Liver Cancer

Overview

Transarterial radioembolization (TARE) is a key treatment option for patients with unresectable hepatocellular carcinoma (HCC), a primary form of liver cancer. TARE is a minimally invasive therapy in which radioactive microspheres are delivered through a microcatheter near the tumour into the liver's blood vessels. Although TARE can significantly improve survival, treatment outcomes remain variable and difficult to predict, mainly because of complex liver vasculature and unpredictable distribution of radioactive microspheres due to uncertain parameters such as catheter tip location, catheter orientation, and injection velocity. The long-term goal would be to make these treatments more predictable and effective by developing a patient-specific pre-treatment planning platform. Blood flow and microsphere transport will be modelled in a digital model of the patient-specific hepatic arterial tree (based on clinical imaging) using computational fluid dynamics (CFD), in combination with Monte Carlo-based radiation dosimetry. Using CFD simulations, we will investigate how variations in treatment parameters influence the microsphere distribution, aiming to better understand their role in treatment variability. This will allow us to predict the dose distribution of a certain treatment and determine potentially a more optimal set of treatment parameters. This research contributes to the broader field of cancer research by laying the foundations for a digital tool for personalized pre-treatment planning. The insights gained could support interventional radiologists in optimizing treatment planning, improving tumor targeting, and minimizing radiation exposure to healthy liver tissue in future TARE procedures.

TARE has become an established locoregional treatment for HCC, the most common form of primary liver cancer. During TARE, radioactive microspheres are injected through a microcatheter in the hepatic arteries to irradiate the tumor from within. The minimally invasive nature of this therapy makes it an important option for patients who are ineligible for surgery. However, despite its clinical relevance, treatment success remains variable between patients. A challenge in TARE is the limited ability to predict how the injected microspheres will spread throughout the arterial network of the liver. As a result, microspheres do not always deposit homogenously within the tumor, and part of the injected dose may end up in healthy liver tissue. This may reduce treatment efficacy while increasing the risk of complications. Current pre-treatment planning and dosimetry approaches are not completely able to predict the resulting radiation dose. The goal of this study is to develop a methodology that enables better understanding and prediction of microsphere distribution and radiation dose during TARE using patient-specific 3D models. These models will be constructed from clinical imaging data such as CT, MRI, PET, SPECT, or DSA, and will be used to analyze hepatic arterial anatomy and dose deposition patterns. By studying these relationships, the project aims to identify which anatomical and procedural factors influence microsphere transport and dose delivery, and how these insights can contribute to improved pre-treatment planning. The clinical part of the study consists solely of collecting retrospective and prospective imaging and treatment data from patients undergoing TARE, without altering clinical workflow. Therefore, no additional risk or discomfort is introduced for participating patients. The collected data will serve as the foundation for developing and validating the 3D models. In the long term, the resulting insights may help optimize catheter positioning, injection parameters, and dose planning, with the ambition of achieving more effective, personalized, and safer TARE treatments.