Federal Grant Award Summary

The National Cancer Institute (NCI) awarded a $414,402 Project Grant to The Regents of the University of California, San Francisco under the Cancer Treatment Research program (CFDA 93.395) for the period May 1, 2026 through April 30, 2028. This research project develops multi-modality imaging and biomarker integration methods to improve glioblastoma radiation therapy targeting. The deliverables include: (1) white matter infiltrative risk maps derived from diffusion tensor imaging (DTI) data that identify population-level tumor infiltration patterns across major white matter tracts and patient-specific fiber density profiles; (2) a transformer-based deep learning framework with cross-attention mechanisms that integrates DTI-based risk maps with anatomical magnetic resonance imaging (MRI), MGMT methylation status, and TERT promoter mutation data; and (3) clinical target volume (CTV) recommendations for personalized radiation therapy planning that account for patient-specific tumor biology and diffuse infiltrative spread patterns.

The research utilizes a dataset exceeding 500 patients with pre-operative DTI imaging, anatomical MRI, and molecular biomarker data to develop and validate improved glioblastoma progression prediction models. By moving beyond conventional uniform geometric expansions around MRI-visible tumor boundaries, the project aims to enhance radiation therapy precision, reduce treatment-induced toxicity, and ultimately improve outcomes for glioblastoma patients, a population currently facing median survival rates of 12-15 months despite aggressive multimodal therapy.