2946 - End-to-End Validation of an Online Adaptive Radiotherapy Workflow using a Dynamic Anthropomorphic Thoraco-Abdominal Phantom
Presenter(s)
A. Bakhtiari Moghaddam1, R. Figueiredo Augusto2, R. Armin3, O. Jaekel4,5, P. Darremont2, G. Echner2, W. Johnen2, P. Haering2, C. Lang2, M. Lifferth2, A. Qubala6, and C. Karger7,8; 1DKFZ, Heielberg, Germany, 2German Cancer Research Center (DKFZ), Heidelberg, Germany, 3German Cancer Research Center (DKFZ) Heidelberg, Germany, Heidelberg, Germany, 4Heidelberg University, Heidelberg 69120, Germany, 5Division of Medical Physics in Radiation Oncology, German Cancer Research Center (DKFZ), Heidelberg, Germany, Heidelberg, Germany, 6Heidelberg Ion Beam Therapy Center (HIT), University Hospital Heidelberg, Germany, Heidelberg, Germany, 7National Center for Radiation Research in Oncology (NCRO), Heidelberg Institute for Radiation Oncology (HIRO), Heidelberg, Germany, Heidelberg, Germany, 8Department of Medical Physics in Radiation Oncology, German Cancer Research Center (DKFZ), Heidelberg, Germany
Purpose/Objective(s): Online adaptive radiotherapy (ART) requires comprehensive end-to-end (E2E) validation tools capable of reproducing organ motion, anatomical deformation, and multimodal imaging characteristics. We developed a modular, dynamic anthropomorphic phantom that allows verification of online adaptive IMRT workflows under realistic thoraco-abdominal motion and deformation scenarios. This Thoraco-abdominal Anthropomorphic Phantom with Motion for Adaptive Radiotherapy (TAM-Ara) was used to perform an E2E test for liver ART.
Materials/Methods: TAM-ARa was constructed using anatomically realistic bone, lung, and abdominal organ models fabricated from tissue-equivalent materials. The modular design enabled simulation of interfractional and intrafractional variations, including ventilator-driven respiratory motion, abdominal deformation via exchangeable deformation modules, and variable gastric filling. Imaging performance was assessed using CT, CBCT, and 3T MRI. E2E tests of an online adaptive IMRT workflow for liver tumors were performed on a Varian Ethos system for several scenarios: (1) static reference configuration, (2) two abdominal and one thoracic deformation scenarios for ionization chamber (IC) dosimetry, and (3) static and abdominal deformation scenarios for radiochromic film measurements. Primary endpoints was the dose deviation between measured and planned dose as well as geometric accuracy of target coverage after adaptation. Secondary endpoints included reproducibility of deformation and feasibility of clinical workflow integration.
Results: The phantom demonstrated realistic radiological properties, with CT Hounsfield units and MRI relaxation parameters comparable to reported in vivo values. Abdominal deformations were reproducible and allowed rapid setup and dosimeter exchange. In E2E adaptive IMRT tests, online plan adaptation restored accurate dose delivery under deformation conditions. IC measurements showed dose deviations below 3% after adaptation. Independent film dosimetry confirmed sub-millimeter geometric accuracy and full planning target volume coverage following adaptive replanning. Without adaptation, deformation-induced deviations were clinically relevant, confirming the necessity of adaptive workflows.
Conclusion: TAM-ARa provides a reproducible, anatomically realistic, and multimodality-compatible platform for comprehensive E2E validation of online adaptive radiotherapy workflows. The phantom enables clinically meaningful assessment of deformation compensation and dose restoration in liver ART and supports future investigations in motion management, dose accumulation, and adaptive workflow optimization.