Personalizing Treatment for Glioblastoma
Chaitra Badve, MD, Serah Choi, MD, PhD, Dan Ma, PhD, and Yong Chen, PhD, are combining Magnetic Resonance Fingerprinting (MRF) with artificial intelligence (AI) based neural networks to improve target delineation in post-operative radiotherapy for glioblastoma. An aggressive brain cancer, glioblastoma is typically treated with surgery to remove as much of the tumor as possible followed by radiotherapy to kill any remaining cancer cells. Radiotherapy helps control local recurrence but carries the risk of neurological deterioration, especially as more of the brain is treated. Current MRI methods cannot clearly distinguish where the tumor has infiltrated the brain beyond the part removed during surgery, known as the peritumoral region, and new, more accurate imaging techniques are needed to improve outcomes for patients with glioblastoma.
Dr. Badve, Dr. Choi, Dr. Ma and Dr. Chen are members of the research team at Case Western Reserve University and University Hospitals Cleveland Medical Center that developed MRF, a novel quantitative imaging tool that that allows for accurate and reproducible high-resolution measurement of multiple tissue properties in a single five-minute scan. Prior studies have shown that MRF is better than conventional MRI for brain imaging setup for radiotherapy. With support from ROI, the team is combining MRF with AI to map tumor infiltration in glioblastoma more accurately for each patient. They are:
Developing an AI-based tool that generates peritumoral infiltration prediction maps by analyzing a retrospective database of MRF images of various newly diagnosed brain tumors to characterize quantitative differences in tissue types and tumor regions.
Validating and optimizing the peritumoral infiltration prediction tool in a prospective study of patients with newly diagnosed glioblastoma.
Assessing the theoretical feasibility of using MRF infiltration prediction maps to personalize radiotherapy treatment planning by comparing treatment volumes and tumor recurrence sites for MRF and standard MRI.
Through this project, Dr. Badve, Dr. Choi, Dr. Ma and Dr. Chen aim to show that MRF-based infiltration prediction maps can be used in radiotherapy treatment planning for glioblastoma to guide target volumes that will allow for increasing the dose to specific regions while lowering the overall brain volume treated. They are developing this new technique to work within current radiology and radiotherapy infrastructure to facilitate wide implementation, and one of their next steps would be to conduct multicenter trials. If successful, this new MRF-based method will support creating personalized radiation treatment strategies that better target tumor cells and reduce neurological deterioration for patients with glioblastoma.