ESTRO 2024 - Abstract Book

S4803

Physics - Quality assurance and auditing

ESTRO 2024

2.Mancosu P, Navarria P, Castagna L, Reggiori G, Stravato A, Gaudino A, Sarina B, Tomatis S, Scorsetti M. Plan robustness in field junction region from arcs with different patient orientation in total marrow irradiation with VMAT. Phys Med. 2015 Nov;31(7):677-82. doi: 10.1016/j.ejmp.2015.05.012. Epub 2015 Jun 9. PMID: 26068115.

1203

Digital Poster

Evaluation of secondary dose calculation as plan-specific QA in online adaptive radiotherapy

Gary Razinskas, Robert Schindhelm, Stefan Weick, Kathrin Breuer, Marjan Nejad, Jörg Tamihardja, Marcus Zimmermann, Andrea Wittig-Sauerwein, Sonja Wegener

University Hospital Würzburg, Department of Radiation Oncology, Würzburg, Germany

Purpose/Objective:

Cone Beam Computed Tomography (CBCT)-based online adaptive radiotherapy (oART) represents an advancing technology, which the Ethos system (Varian Medical Systems) enables through artificial intelligence-assisted adaptation of treatment plans to the patient's daily anatomy. In this context, secondary dose calculation (SDC) with an independent algorithm is the primary method for plan-specific quality assurance (QA) during oART. This study explores the performance observed in our institution regarding the vendor’s proprietary SDC solution, with the aim of establishing thresholds for informed decision-making during oART, thus ensuring accurate and safe treatment.

Material/Methods:

The adaptive treatment focused on patients diagnosed with cancer in the pelvic region, encompassing patient cases with prostate, rectum, and cervix cancer, and employing multi-field intensity-modulated radiotherapy (IMRT) plans. Treatment planning was performed in the Ethos Treatment Management software (version 2.1, Varian Medical Systems), while the actual treatment sessions were conducted using an Ethos linac (Halcyon version 3.1, Varian Medical Systems). Plan-specific QA was performed using SDC within the Mobius3D platform (version 4.0.2, Varian Medical Systems), utilizing its Collapsed Cone algorithm. To optimize the Dosimetric Leaf Gap (DLG) correction factor, the initial 15 adaptive reference plans were utilized. Dose measurements within the Mobius Verification Phantom were conducted using a PinPoint chamber placed in a position of homogeneous dose. Consistency between dose distributions from Ethos and Mobius3D was assessed for subsequent 235 adaptive sessions. This assessment relied on a global gamma analysis employing a 3%/2mm criterion (10% low dose threshold), encompassing both the overall volume and planning target volume (PTV). Moreover, deviations in dose-volume histogram (DVH) parameters, specifically D mean and D 95% , were scrutinized for the target volumes.

Results:

Initially, the uncorrected golden beam calculation model in Mobius3D exhibited a systematic global dose deviation of -2.1% compared to the measured values averaged over the 15 adaptive reference plans. With the determined adjustment of the DLG correction factor for IMRT to a value of 0.6 to fit the measurement results, enhancements in mean gamma pass rates were achieved. Notably, for the unrestricted volume the gamma pass rates increased from