Taschereau R, Roy J, Pouliot J
Radiation Oncology Service, Centre de Recherche en Cancérologie de l'Université Laval, Canada.
Med Phys. 1999 Sep;26(9):1952-9. doi: 10.1118/1.598700.
The objective of this study is to use Monte Carlo simulations to assess the sensitivity of implant planning methods to seed misplacement. A model of seed misplacement is first developed. It is based upon data gathered after a study on source migration performed on 30 patients treated with I-125 transperineal implants. It consists of applying elementary transformations to every needle in a loading plan to produce a distorted implant mimicking the effect of migration. After being validated, the model has been used to tune the inverse planning system in use at our institution. The new planning system is now used clinically and actual results are compared with those predicted by simulations. Simulations were also used to compare our planning method with others. The new planning system increased the average postimplant dose-volume histogram DVH(160) from 82% to 93%, which is the value predicted by the simulations. This improvement is due to an increased dose margin providing coverage even in the presence of migration. At the same time, the dose to the urethra remained at 267 Gy because of a special protection feature included in the planning system. Some other implant planning methods are not as robust [average DVH(160) ranging from 76% to 85%] and deliver a higher dose to the urethra (close to 400 Gy). To conclude, a simple model of source migration can provide realistic feedback about sensitivity to migration of planning methods. It allowed a significant clinical improvement at our institution. The improved inverse planning system provided better coverage with fewer seeds (but equal total activity) than a manual method. Hence, a properly tuned inverse planning system has the potential to deliver the less sensitive plans. The model also helped demonstrate that planning methods are not equally robust to migration and that they should not be evaluated solely by the plans they produce, but also by their clinical (or simulated) results.
本研究的目的是使用蒙特卡罗模拟来评估植入计划方法对籽源误置的敏感性。首先建立了一个籽源误置模型。该模型基于对30例接受I - 125经会阴植入治疗的患者进行源迁移研究后收集的数据。它包括对加载计划中的每根针应用基本变换，以产生一个模拟迁移效应的扭曲植入物。经验证后，该模型已用于调整我们机构使用的逆向计划系统。新的计划系统现已用于临床，并将实际结果与模拟预测结果进行比较。模拟还用于将我们的计划方法与其他方法进行比较。新的计划系统将植入后平均剂量体积直方图DVH(160)从82%提高到了93%，这与模拟预测的值一致。这种改善归因于剂量裕度的增加，即使在存在迁移的情况下也能提供覆盖。同时，由于计划系统中包含的特殊保护功能，尿道所受剂量保持在267 Gy。其他一些植入计划方法则不那么稳健（平均DVH(160)在76%至85%之间），并且会给尿道带来更高的剂量（接近400 Gy）。总之，一个简单的源迁移模型可以提供关于计划方法对迁移敏感性的真实反馈。它使我们机构在临床上有了显著改善。改进后的逆向计划系统比手动方法能用更少的籽源（但总活度相同）提供更好的覆盖。因此，一个经过适当调整的逆向计划系统有可能提供对迁移不太敏感的计划。该模型还有助于证明计划方法对迁移的稳健性并不相同，并且不应仅根据它们生成的计划来评估，还应根据其临床（或模拟）结果来评估。
