Department of Radiation Oncology, Erasmus-MC Daniel den Hoed Cancer Center, Rotterdam, the Netherlands.
Int J Radiat Oncol Biol Phys. 2012 Aug 1;83(5):1617-23. doi: 10.1016/j.ijrobp.2011.10.011. Epub 2012 Jan 21.
To design and evaluate individualized nonadaptive and online-adaptive strategies based on a pretreatment established motion model for the highly deformable target volume in cervical cancer patients.
For 14 patients, nine to ten variable bladder filling computed tomography (CT) scans were acquired at pretreatment and after 40 Gy. Individualized model-based internal target volumes (mbITVs) accounting for the cervix and uterus motion due to bladder volume changes were generated by using a motion-model constructed from two pretreatment CT scans (full and empty bladder). Two individualized strategies were designed: a nonadaptive strategy, using an mbITV accounting for the full-range of bladder volume changes throughout the treatment; and an online-adaptive strategy, using mbITVs of bladder volume subranges to construct a library of plans. The latter adapts the treatment online by selecting the plan-of-the-day from the library based on the measured bladder volume. The individualized strategies were evaluated by the seven to eight CT scans not used for mbITVs construction, and compared with a population-based approach. Geometric uniform margins around planning cervix-uterus and mbITVs were determined to ensure adequate coverage. For each strategy, the percentage of the cervix-uterus, bladder, and rectum volumes inside the planning target volume (PTV), and the clinical target volume (CTV)-to-PTV volume (volume difference between PTV and CTV) were calculated.
The margin for the population-based approach was 38 mm and for the individualized strategies was 7 to 10 mm. Compared with the population-based approach, the individualized nonadaptive strategy decreased the CTV-to-PTV volume by 48% ± 6% and the percentage of bladder and rectum inside the PTV by 5% to 45% and 26% to 74% (p < 0.001), respectively. Replacing the individualized nonadaptive strategy by an online-adaptive, two-plan library further decreased the percentage of bladder and rectum inside the PTV (0% to 10% and -1% to 9%; p < 0.004) and the CTV-to-PTV volume (4-96 ml).
Compared with population-based margins, an individualized PTV results in better organ-at-risk sparing. Online-adaptive radiotherapy further improves organ-at-risk sparing.
设计并评估基于宫颈癌患者治疗前建立的运动模型的个体化非自适应和在线自适应策略，以适应该高度变形的靶区。
对 14 名患者，在治疗前和 40 Gy 后，采集了 9 到 10 个可变膀胱充盈的 CT 扫描。通过使用从两个治疗前 CT 扫描（全膀胱充盈和排空）构建的运动模型，生成了个体化基于模型的内部靶区（mbITV），该模型考虑了由于膀胱体积变化引起的宫颈和子宫运动。设计了两种个体化策略：一种是非自适应策略，使用整个治疗过程中膀胱体积变化的全范围的 mbITV；另一种是在线自适应策略，使用膀胱体积子范围的 mbITV 构建计划库。后者通过基于测量的膀胱体积从库中选择当日的计划来在线调整治疗。使用未用于 mbITV 构建的 7 到 8 次 CT 扫描评估个体化策略，并与基于人群的方法进行比较。确定了围绕计划宫颈-子宫和 mbITV 的几何均匀边界，以确保足够的覆盖范围。对于每种策略，计算了宫颈-子宫、膀胱和直肠体积在计划靶区（PTV）内的百分比，以及临床靶区（CTV）与 PTV 之间的体积（PTV 和 CTV 之间的体积差异）。
基于人群的方法的边界为 38 mm，个体化策略的边界为 7 到 10 mm。与基于人群的方法相比，个体化非自适应策略使 CTV 与 PTV 之间的体积减少了 48% ± 6%，使膀胱和直肠在 PTV 内的百分比分别减少了 5%至 45%和 26%至 74%（p < 0.001）。用在线自适应、两计划库代替个体化非自适应策略，进一步减少了 PTV 内膀胱和直肠的百分比（0%至 10%和-1%至 9%；p < 0.004）和 CTV 与 PTV 之间的体积（4-96 ml）。
与基于人群的边界相比，个体化 PTV 可更好地保护危及器官。在线自适应放疗进一步改善了危及器官的保护。
