E = mc²: Environment-driven Mathematical Modeling for the physical microenvironment in clinical cancer imaging
Project PI: Kristin R. Swanson, Ph.D.
Gliomas are uniformly fatal primary brain tumors with a unique ability to interact with the tissue microenvironment as a result of their single cell diffuse invasion of otherwise normal brain tissue peripheral to the frank abnormality. Building upon our microenvironment-driven model for glioma proliferation and invasion (PIHNA model), we are developing a technique for predicting patient-specific maps of hypoxia that are directly comparable to patient PET images using the hypoxia radiotracer [18F]-fluoromisonidazole.
- By incorporating the effects of the pharmacokinetics of the FMISO as well as the complex physics involved in the image acquisition and reconstruction process of the PET, we have been able to successfully predict patient-specific FMISO-PET as well as a continuous map of hypoxia is generated that does not have the noisy disadvantages associated with PET imaging.
- Following the Swanson Lab’s decade of experience with parametrizing patient-specific models using actual patient MRIs as input, the only data need to inform the PIHNA model parameters needed to produce the simulated FMISO-PET come from routinely available serial pre-treatment MRIs.
- This tool allows for a direct quantification of the in vivo connection between kinetics of tumor aggressiveness (quantified by the patient-specific mathematical model) and physical mE elements, such as hypoxia, that are routinely only available through poorly resolved imaging such as FMISO-PET.
- First manuscript submitted in Oct 09: S. Gu, G. Chakraborty, K. Champley, A. Alessio, J. Claridge, R. Rockne, M. Muzi, K. A. Krohn, A. M. Spence, E. C. Alvord Jr, A. R. A. Anderson, P. Kinahan, K. R. Swanson. Applying A Patient-Specific Bio-Mathematical Model of Glioma Growth to Develop Virtual [18F]-FMISO PET Images