Radiation and medical oncologists often face a critical question when treating patients with glioblastoma, the most aggressive form of brain cancer: is it true progression or pseudoprogression? Distinguishing between these types of progression is crucial for treatment management.
Eric Mellon, MD, PhD, a radiation oncologist and researcher at the Sylvester Comprehensive Cancer Center, presented the results of two recent studies on this topic at the annual meeting of the American Society for Radiation Oncology. The first study focused on monitoring tumor changes during radiation therapy to predict treatment response. Using MRI-guided linear accelerator systems, the researchers were able to detect patterns within these volume changes.
They trained a model to predict between no progression, pseudoprogression, and true progression, achieving an overall accuracy of 86%. This model could be an early marker of treatment response. The researchers are now looking to automate real-time tumor monitoring using deep learning.
In the second study, the researchers analyzed delta-radiomic features extracted from MR-linac systems to predict true progression or pseudoprogression. They found that changes in the tumor microenvironment occurred within the first 15 treatment sessions, suggesting that early interventions may help poorly responding patients.
The findings of both studies support the use of MR-linac systems for predicting treatment response during therapy and call for further research with larger patient cohorts.
Sources:
– University of Miami Miller School of Medicine