Development and preclinical validation of magnetomotive embolization for enhanced targeting of hepatocellular carcinoma

Background: Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer and remains a leading cause of cancer mortality worldwide. Catheter-directed transarterial embolization (with or without chemotherapy) is the standard treatment for intermediate-stage HCC, but its survival benefit remains modest, reflecting limitations such as incomplete tumor embolization. To address these limitations, we aimed to develop an embolization approach that improves intratumoral bead localization and tumor embolization through biocompatible magnetic embolization beads and magnetic field guidance. Methods : A dual-cylinder NdFeB magnet system was engineered to generate focused magnetic fields up to 25 mT at 7.5 cm from the magnet surface, creating a stable magnetic trap. Polycaprolactone-based magnetic embolization beads (40-60 µm) incorporating ~30 nm Fe3O4 iron oxide nanoparticles were synthesized using emulsion polymerization and characterized for morphology, uniformity, and magnetization. Magnet-guided bead delivery was assessed in 3D-printed vascular tumor phantoms and validated in vivo using the rabbit VX2 liver tumor model. Tumor-bearing rabbits (n=12) underwent transarterial embolization and were randomized into magnet-guided (5 Hz alternating rotation, 15 minutes) or control groups (n=6 each). Results : Microscopic analysis confirmed uniform spherical bead morphology with a narrow size distribution, while SQUID magnetometry demonstrated ferrimagnetic behavior of Fe3O4 nanoparticle cores (magnetization 63.6 emu/g, coercivity 20 Oe at 800 kA/m), confirming strong magnetic responsiveness, and supporting MR-safety. Powder X-ray diffractometry confirmed Fe3O4 with no anomalous peaks suggesting the presence of undesirable iron oxide species. In tumor phantoms, the magnet system successfully concentrated magnetic beads within targeted regions at preclinical relevant distances, achieving reproducible spatial control and contrasting with heterogeneous dispersion in controls. In vivo, transarterial delivery was technically feasible and well tolerated in all tumor-bearing rabbits with no acute complications. Histopathologic analysis demonstrated concentration of the magnetic beads in tumor vasculature in magnet-assisted animals. No device-related toxicity was observed. Conclusion : This study establishes proof-of-concept for magnetomotive embolization as a novel and feasible locoregional therapy for HCC. The findings demonstrate technical feasibility, procedural safety, and effective intratumoral bead localization in both phantom and preclinical animal models. Building on these results, future work will focus on integrating robotic magnetic field control for dynamic targeting and performing quantitative efficacy and safety studies in large-animal models to advance this technology toward clinical translation.