Department of Radiation Oncology, ErasmusMC - Daniel den Hoed Cancer Center, Rotterdam, The Netherlands.
Int J Radiat Oncol Biol Phys. 2013 Oct 1;87(2):401-6. doi: 10.1016/j.ijrobp.2013.06.2032.
To investigate interfraction setup variations of the primary tumor, elective nodes, and vertebrae in laryngeal cancer patients and to validate protocols for cone beam computed tomography (CBCT)-guided correction.
For 30 patients, CBCT-measured displacements in fractionated treatments were used to investigate population setup errors and to simulate residual setup errors for the no action level (NAL) offline protocol, the extended NAL (eNAL) protocol, and daily CBCT acquisition with online analysis and repositioning.
Without corrections, 12 of 26 patients treated with radical radiation therapy would have experienced a gradual change (time trend) in primary tumor setup ≥4 mm in the craniocaudal (CC) direction during the fractionated treatment (11/12 in caudal direction, maximum 11 mm). Due to these trends, correction of primary tumor displacements with NAL resulted in large residual CC errors (required margin 6.7 mm). With the weekly correction vector adjustments in eNAL, the trends could be largely compensated (CC margin 3.5 mm). Correlation between movements of the primary and nodal clinical target volumes (CTVs) in the CC direction was poor (r(2)=0.15). Therefore, even with online setup corrections of the primary CTV, the required CC margin for the nodal CTV was as large as 6.8 mm. Also for the vertebrae, large time trends were observed for some patients. Because of poor CC correlation (r(2)=0.19) between displacements of the primary CTV and the vertebrae, even with daily online repositioning of the vertebrae, the required CC margin around the primary CTV was 6.9 mm.
Laryngeal cancer patients showed substantial interfraction setup variations, including large time trends, and poor CC correlation between primary tumor displacements and motion of the nodes and vertebrae (internal tumor motion). These trends and nonrigid anatomy variations have to be considered in the choice of setup verification protocol and planning target volume margins. eNAL could largely compensate time trends with minor prolongation of fraction time.
研究喉癌患者原发肿瘤、选择性淋巴结和椎体的分次间摆位变化，并验证锥形束 CT（CBCT）引导校正方案。
对 30 例患者进行研究，通过分次治疗中 CBCT 测量的位移来研究人群摆位误差，并模拟无行动水平（NAL）离线方案、扩展 NAL（eNAL）方案以及每日 CBCT 采集与在线分析和重新定位的残余摆位误差。
未经校正，26 例接受根治性放疗的患者中有 12 例（11/12 例在尾端方向，最大 11 毫米）在分次治疗期间原发肿瘤的头脚方向（CC 方向）会出现逐渐变化（时间趋势）。由于这些趋势，使用 NAL 校正原发肿瘤位移会导致较大的 CC 残余误差（需要的边缘 6.7 毫米）。在 eNAL 中，通过每周校正矢量调整，可以在很大程度上补偿这些趋势（CC 边缘 3.5 毫米）。在 CC 方向上，原发肿瘤和淋巴结临床靶区（CTV）的运动之间相关性较差（r(2)=0.15）。因此，即使对原发 CTV 进行在线摆位校正，淋巴结 CTV 的 CC 边缘也需要 6.8 毫米。对于椎体，一些患者也观察到较大的时间趋势。由于原发 CTV 与椎体之间 CC 相关性较差（r(2)=0.19），即使每天在线重新定位椎体，原发 CTV 周围也需要 6.9 毫米的 CC 边缘。
喉癌患者表现出明显的分次间摆位变化，包括大的时间趋势，以及原发肿瘤位移与淋巴结和椎体运动之间的 CC 相关性较差（内部肿瘤运动）。这些趋势和非刚性解剖结构变化必须在选择摆位验证方案和计划靶区边缘时加以考虑。eNAL 可以通过轻微延长分次时间来大大补偿时间趋势。
