Department of Radiation Oncology, University of Wurzburg, Würzburg, Germany.
Med Phys. 2012 Feb;39(2):713-20. doi: 10.1118/1.3671906.
For fast adaptation of step and shoot intensity modulated radiotherapy (IMRT) plans, monitor units (MU)-preserving methods which modify only the segment shapes have been proposed in the literature. In this work, two such adaptation methods are applied to intensity modulated arc therapy (IMAT) and their results are compared to that of a newly optimized IMAT plan.
In a simplified cylindrically symmetric model, the organ at risk (OAR) is surrounded by the planning target volume (PTV). For the initial plan, a steep dose gradient is produced by variants of double arc (IMAT) plans. To simulate situations which require adaptation, the OAR radius and the inner PTV radius have been varied. One adaptation method (Warp) is based on a mesh spanned over structures identified within the beam's eye view (BEV). Changes to the structure projections warp the mesh. For the adaptation, the segment shapes are fixed to the mesh. The other method (2-Step) uses geometrical 3D information from the computed tomography (CT). For comparison, the objective function representing the dose to the PTV as well as the mean and the maximum dose to the OAR is used.
For the narrow segments that compensate the underdosage in the PTV areas proximate to the OAR, the Warp method suggests contrary adaptation rules compared to the 2-Step method. In contrast to Warp, the 2-Step method approximates the behavior of a newly optimized plan and leads to better dose homogeneity in the clinical target volume (CTV) and the PTV, whilst simultaneously sparing the OAR.
For minor changes associated with less steep dose gradients, both Warp and 2-Step methods are suitable. However, the 2-Step method should be preferred for more challenging cases, where steep dose gradients between the OAR and the concave PTV are needed. For considerable interfractional reductions of the gap between the OAR and the PTV, where especially steep dose gradients have to be generated, MU-preserving adaptation techniques are not adequate. In this case, narrower segments in the initial plan can be used to facilitate the adaptation. Otherwise, non-MU-preserving adaptation methods have to be applied. Further work is needed to include clinical cases with more complex geometries and expand the methods to IMRT techniques.
为了快速适应步进式和扫射强度调制放疗（IMRT）计划，文献中提出了仅修改射野内子野形状的保持 MU 量的方法。本研究将两种适应方法应用于强度调制弧形治疗（IMAT），并将其结果与新优化的 IMAT 计划进行比较。
在简化的圆柱对称模型中，危及器官（OAR）被规划靶区（PTV）包围。对于初始计划，通过双弧（IMAT）计划的变体产生陡峭的剂量梯度。为了模拟需要适应的情况，已经改变了 OAR 半径和内 PTV 半径。一种适应方法（变形）基于在射野影像（BEV）中识别的结构上展开的网格。结构投影的变化会使网格变形。对于适应，子野形状固定在网格上。另一种方法（两步法）使用来自计算机断层扫描（CT）的几何 3D 信息。为了比较，使用代表 PTV 剂量的目标函数以及 OAR 的平均和最大剂量。
对于补偿 OAR 附近 PTV 区域欠剂量的窄射野，变形方法与两步法建议的适应规则相反。与变形法相比，两步法近似于新优化计划的行为，在同时保护 OAR 的同时，导致临床靶区（CTV）和 PTV 内的剂量均匀性更好。
对于与陡峭剂量梯度相关的较小变化，变形法和两步法都适用。然而，对于需要 OAR 和凹形 PTV 之间陡峭剂量梯度的更具挑战性的情况，应优先选择两步法。对于 OAR 和 PTV 之间间隙的明显分次减少，需要生成特别陡峭的剂量梯度，保持 MU 的适应技术是不够的。在这种情况下，可以在初始计划中使用更窄的射野来方便适应。否则，必须应用非 MU 保持的适应方法。需要进一步的工作来包括具有更复杂几何形状的临床病例，并将方法扩展到 IMRT 技术。
