Vrije Universiteit Brussel (VUB), Department of Electronics and Informatics (ETRO), Pleinlaan 2, B-1050 Brussels, Belgium; imec, Kapeldreef 75, B-3001 Leuven, Belgium; Vrije Universiteit Brussel (VUB), Faculty of Medicine and Farmacy, Brussels, Belgium; Department of Radiotherapy, Universitair Ziekenhuis Brussel, Brussels, Belgium.
Vrije Universiteit Brussel (VUB), Department of Electronics and Informatics (ETRO), Pleinlaan 2, B-1050 Brussels, Belgium; imec, Kapeldreef 75, B-3001 Leuven, Belgium.
Radiother Oncol. 2018 Feb;126(2):339-346. doi: 10.1016/j.radonc.2017.09.001. Epub 2017 Oct 6.
To evaluate the short and long-term variability of breathing induced tumor motion.
3D tumor motion of 19 lung and 18 liver lesions captured over the course of an SBRT treatment were evaluated and compared to the motion on 4D-CT. An implanted fiducial could be used for unambiguous motion information. Fast orthogonal fluoroscopy (FF) sequences, included in the treatment workflow, were used to evaluate motion during treatment. Several motion parameters were compared between different FF sequences from the same fraction to evaluate the intrafraction variability. To assess interfraction variability, amplitude and hysteresis were compared between fractions and with the 3D tumor motion registered by 4D-CT. Population based margins, necessary on top of the ITV to capture all motion variability, were calculated based on the motion captured during treatment.
Baseline drift in the cranio-caudal (CC) or anterior-poster (AP) direction is significant (ie. >5 mm) for a large group of patients, in contrary to intrafraction amplitude and hysteresis variability. However, a correlation between intrafraction amplitude variability and mean motion amplitude was found (Pearson's correlation coefficient, r = 0.72, p < 10). Interfraction variability in amplitude is significant for 46% of all lesions. As such, 4D-CT accurately captures the motion during treatment for some fractions but not for all. Accounting for motion variability during treatment increases the PTV margins in all directions, most significantly in CC from 5 mm to 13.7 mm for lung and 8.0 mm for liver.
Both short-term and day-to-day tumor motion variability can be significant, especially for lesions moving with amplitudes above 7 mm. Abandoning passive motion management strategies in favor of more active ones is advised.
评估呼吸诱导肿瘤运动的短期和长期变化。
对 19 个肺部和 18 个肝脏病变的 SBRT 治疗过程中的 3D 肿瘤运动进行评估，并与 4D-CT 上的运动进行比较。可以使用植入的基准点提供明确的运动信息。治疗过程中使用快速正交荧光透视（FF）序列来评估运动。对同一分次治疗中不同 FF 序列之间的几个运动参数进行比较，以评估分次内变异性。为了评估分次间变异性，比较了分次间的幅度和滞后与 4D-CT 注册的 3D 肿瘤运动。基于治疗期间捕获的运动，根据人群计算了必要的 ITV 上的边界，以捕获所有运动变异性。
与分次内幅度和滞后变异性相比，很大一部分患者的头脚（CC）或前后（AP）方向的基线漂移显著（即>5mm）。然而，发现分次内幅度变异性与平均运动幅度之间存在相关性（Pearson 相关系数，r=0.72，p<10）。46%的所有病变的幅度变异性都有统计学意义。因此，4D-CT 在某些分次中准确地捕获了治疗期间的运动，但并非所有分次。在所有方向上，考虑到治疗期间的运动变异性会增加 PTV 边界，在 CC 方向上增加最为显著，肺部从 5mm 增加到 13.7mm，肝脏从 8.0mm 增加到 13.7mm。
短期和日常肿瘤运动变异性都可能很显著，特别是对于幅度超过 7mm 的病变。建议放弃被动运动管理策略，转而采用更主动的策略。
