P Inman*, J Houri, J Gounley, G Agasthya, A Kapadia, Oak Ridge National Laboratory, Oak Ridge, TN

• 
  P Inman

WE-B-BRC-2 (Wednesday, 7/13/2022) 8:30 AM - 9:30 AM [Eastern Time (GMT-4)]

Ballroom C

Purpose: We developed an automated computational multiscale framework for precision dosimetry to track radiation effects at the cellular to whole-body scale. This framework’s capabilities were demonstrated using external-beam fractionated proton therapy for treating breast cancer tumors.

Methods: Our framework consists of a parallelizable automated pipeline that combines multicellular phenomenological models for tumor growth, extended cardiac-torso (XCAT) phantoms for organ-level precision, and GEANT4 to track radiation dose. Multicell tumor models are generated in Compucell3D (CC3D) and placed within XCAT phantoms for irradiation using GEANT4. Cell doses are tracked to estimate tumor cell survival in CC3D using linear-quadratic models, and voxelized organ dose distributions are recorded to estimate dose to surrounding tissue. We demonstrated the utility of this framework using a validated breast cancer monolayer model and tracked tumor evolution in 4 cases: no treatment and three external-beam 50 MeV proton doses: single 4-Gy fraction early in tumor development, single 20-Gy fraction late in tumor development, and a 40-Gy cumulative dose delivered over 10 fractions.

Results: The framework predicted cell growth and response to treatment consistent with expected estimates as well as previously published results from literature. The automated communication pipeline between CC3D and GEANT4 enabled seamless estimation of dose maps and corresponding tumor response at each desired time point. Proportional dose-dependent decreases in cell volume were observed following each irradiation (single or fractionated). The untreated simulation was executed in 7 minutes, while each fraction of 2.5E6 proton histories required 1.5 hours.

Conclusion: We developed and tested a computational framework using CC3D and GEANT4 to iteratively evaluate radiation effects on multicellular models embedded within XCAT phantoms. Future work includes validating this framework and expanding it to include brachytherapy, radioisotope therapy, and other cancer types to predict cell survival and treatment outcomes in a variety of precision oncology applications.

Funding Support, Disclosures, and Conflict of Interest: This work was supported by the Office of Biological and Environmental Research, Biological Systems Science Division and Laboratory Directed Research and Development Program of ORNL, managed by UT-Battelle, LLC, for the U.S. DOE. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. DOE.

Keywords

Radiation Therapy, Microdosimetry, Monte Carlo

Taxonomy

TH- Response Assessment: Modeling: other than machine learning

Contact Email