Room: AAPM ePoster Library
Purpose: Create a patient-specific whole-body computational phantom to evaluate out-of-field organ exposures for the investigation of secondary malignancy risk due to neutron dose from proton therapy.
Methods: Whole-body CT (WBCT) scans from patients undergoing total body photon irradiation were retrospectively collected as reference. A hybrid phantom was created using an excised portion of one of the WBCT scans for the in-field region, an adult mesh-type reference computational phantom (MRCP) scaled to patient measurements (taken from the torso) for the far out-of-field regions, and a blended transition (near-field) between the two. A representative treatment plan was created on the in-field CT images using a research version of a commercial treatment planning system. The treatment was then simulated with a general-purpose Monte Carlo code on the hybrid phantom, the scaled and unscaled MRCP, and the WBCT. Total and neutron absorbed dose and fluence averaged neutron energy were registered in each voxel using customized scorers, enabling equivalent dose estimates based on ICRP Publication 92. Absorbed dose/equivalent dose distributions in the WBCT were compared to all phantoms.
Results: The hybrid phantom produced a more accurate estimation of the absorbed/equivalent dose in the in-field and out-of-field regions compared to the other phantoms. For the hybrid phantom, the organs in-field had a percent difference for absorbed/equivalent dose of less than 2% when compared to the WBCT. Far out-of-field, the hybrid outperformed the scaled and unscaled MRCP (10.5% average absorbed dose difference vs. 89.6% and 135.7%).
Conclusion: Preliminary results revealed the hybrid phantom performed well even in organs far from the treatment field while retaining the accuracy of the patient CT in the in-field. However, the introduction of the hybrid phantom, although improved over generic phantoms, struggled to estimate dose in the colon due to the complexity and extent of that organ.
Funding Support, Disclosures, and Conflict of Interest: This project was funded by the German Research Foundation (DFG) Graduiertenkolleg 2274.