AI-Guided Arterial Mapping Improves Brain Tumor Drug Delivery

Although intra-arterial therapy can limit systemic exposure, single-artery infusions may miss parts of the lesion and allow off-target delivery. These limitations can reduce dosing efficiency and complicate repeat treatment sessions. To help address this challenge, researchers have developed an artificial intelligence–guided approach that maps a tumor’s arterial supply and directs multi-territory super-selective infusion.

Researchers at The University of Texas MD Anderson Cancer Center (Houston, TX, USA) developed an AI-assisted system to plan intra-arterial therapy for malignant brain tumors. The method analyzes angiographic data to identify tumor-feeding arterial pedicles before treatment. By shifting from a single-artery strategy to a multi-pedicle plan tailored to each patient, the approach is designed to improve tumor coverage and treatment precision. The work was presented at the Society of NeuroInterventional Surgery’s (SNIS) 23rd Annual Meeting.

The technique, described as AI-guided multi-territory super-selective endovascular infusion, begins with algorithmic mapping of all vessels supplying the tumor. Clinicians then confirm the AI output with advanced imaging. Therapy is delivered super-selectively through each mapped artery, with dosing tailored to the vessel’s relative blood supply to the tumor. This approach aims to increase on-target delivery while limiting infusion outside the intended regions.

In an initial series, three patients with malignant brain tumors underwent treatment using this workflow. AI-assisted mapping identified multiple tumor-feeding arteries in every case, enabling treatment of all tumor-feeding pedicles. Estimated coverage exceeded 85% of each tumor with the multi-pedicle approach, compared with less than 65% using single-pedicle proximal infusion. The procedure was also reported to be safely repeatable two weeks later. Further studies are needed to determine whether improved tumor coverage translates into better clinical outcomes.

An abstract of the work appears in the SNIS 23rd Annual Meeting Oral Abstracts.

"One of the biggest challenges in treating malignant brain tumors is that every patient's anatomy is different," said Christopher Young, MD, Ph.D., the study's primary author.

"AI gives us another tool to personalize treatment based on each patient's unique blood supply, with the goal of delivering therapy more precisely. While more research is needed, this approach has the potential to improve patient care and marks an exciting advancement in neurointerventional oncology," said Young. 

Related Links
University of Texas MD Anderson Cancer Center
Society of NeuroInterventional Surgery