AI-Guided Technique Improves Targeted Therapy Delivery for Malignant Brain Tumors
核心洞察
Researchers at MD Anderson Cancer Center developed an AI-assisted technique that identifies all tumor-feeding arteries to improve intra-arterial therapy delivery for malignant brain tumors (搜索).
The multi-pedicle approach achieved more than 85% tumor coverage in all three treated patients, compared to less than 65% with traditional single-artery infusion.
The AI-guided method creates patient-specific maps of tumor blood supply, enabling precise, tailored dosing to each feeding artery while reducing off-target delivery.
A novel artificial intelligence-guided technique that maps the complete blood supply of malignant brain tumors (搜索) could significantly improve the precision of intra-arterial therapy delivery, according to research presented at the Society of NeuroInterventional Surgery's (SNIS) 23rd Annual Meeting.
The approach, developed by researchers at The University of Texas MD Anderson Cancer Center, uses AI to identify all tumor-feeding arterial pedicles before treatment, enabling physicians to deliver therapy through multiple arteries rather than relying on a single-vessel approach.
Addressing a Key Limitation in Intra-Arterial Delivery
Intra-arterial therapy is a minimally invasive procedure in which a catheter delivers medication directly through an artery, allowing for highly targeted dosing while minimizing systemic side effects. However, traditional delivery through a single artery often results in incomplete tumor coverage because many malignant brain tumors (搜索) receive blood from multiple vessels.
"One of the biggest challenges in treating malignant brain tumors (搜索) is that every patient's anatomy is different," said Christopher Young, MD, PhD, 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."
Clinical Results with Multi-Pedicle Infusion
In the study, titled "From Single-Pedicle to Whole Tumor Coverage: AI-guided Multi-territory Super-selective Endovascular Infusion for Brain Tumors," three patients with malignant brain tumors (搜索) underwent treatment using the AI-assisted technique. After the AI system identified tumor-feeding arteries, physicians confirmed the findings with advanced imaging and then delivered a specific amount of therapy tailored to each artery based on the volume of blood each vessel supplied.
The AI-assisted mapping successfully identified multiple tumor-feeding arteries in every patient. The multi-pedicle approach covered more than 85% of each tumor in all cases, compared to less than 65% coverage achieved with single-pedicle proximal infusion. By creating a patient-specific map of the tumor's blood supply, physicians delivered therapy more precisely to the intended areas while reducing off-target delivery.
Researchers also demonstrated that the procedure could be safely repeated two weeks later, suggesting potential for repeated treatment cycles.
Future Directions
While the results represent a promising advancement in neurointerventional oncology, the researchers emphasized that additional studies will be needed to determine whether improved tumor coverage leads to better outcomes for patients. The technique marks a step toward truly personalized endovascular therapy for brain tumors, leveraging AI to account for the unique vascular anatomy of each patient.