ESTRO meets Asia 2024 - Abstract Book

S328

Physics – Motion management and adaptive radiotherapy

ESTRO meets Asia 2024

the workflow in our center. All three plan types utilized the same beam arrangement with two 6 MV partial arcs, and the iso-center was set at the center of the PTV. These plans were optimized with the goal of achieving the clinical objectives outlined in our departmental protocol, which aligns with the recommendations of the QUANTEC review [2,3]. Compared to the NR plans, the RO plans incorporated additional robust optimization objectives that simultaneously accounted for the dose distribution in both the nominal and simulated scenarios. The simulated scenarios replicated various uncertainties, including a scenario with a 5 mm posterior shift of the iso-center to simulate patient setup error, a scenario with a locally deformed CT to simulate a 6 mm anterior swelling of the breast CTV, and a combined scenario incorporating both of these simulated uncertainties. In RayStation, RO was performed using the minimax optimization method, which aimed to ensure sufficient target coverage and maintain tolerable hotspot levels even in the worst-case scenario. The VB plans were optimized with a 10 mm anterior expansion of a pseudo-target on the CTV, along with a 10 mm virtual bolus applied on top of the body surface. The virtual bolus was removed before the final dose calculation and the dose was then normalized to achieve a desired PTV coverage. The plan quality and robustness of the three types of plans were subsequently evaluated by comparing their dose metrics in the nominal and simulated scenarios. In the worst-case scenario, the mean PTV V 95 % of the VB plans (95.1%) was slightly superior to that of the RO plan (94.2%), while both were significantly better than that of the NR plans (84.6%) as shown in figure 1. However, in the nominal scenario, although the mean PTV coverage was similar across three plan types, the mean OARs doses of the VB plans were considerably higher than that of the NR plans by as much as 1.8 Gy, suggesting that the use of virtual bolus degraded the plan quality. In contrast, the mean dose differences for OARs between NR and RO plans were small and below 0.5 Gy for all OARs. The dose metrics of the three plan types in the nominal scenario are presented in Table 1. Results: